This action might not be possible to undo. Are you sure you want to continue?

BooksAudiobooksComicsSheet Music### Categories

### Categories

Scribd Selects Books

Hand-picked favorites from

our editors

our editors

Scribd Selects Audiobooks

Hand-picked favorites from

our editors

our editors

Scribd Selects Comics

Hand-picked favorites from

our editors

our editors

Scribd Selects Sheet Music

Hand-picked favorites from

our editors

our editors

Top Books

What's trending, bestsellers,

award-winners & more

award-winners & more

Top Audiobooks

What's trending, bestsellers,

award-winners & more

award-winners & more

Top Comics

What's trending, bestsellers,

award-winners & more

award-winners & more

Top Sheet Music

What's trending, bestsellers,

award-winners & more

award-winners & more

P. 1

Kaczynski Mischaikow Mrozek_Algebraic Topology - A Computational Approach|Views: 34|Likes: 0

Published by Борис Новиков

See more

See less

https://www.scribd.com/doc/66935407/Kaczynski-Mischaikow-Mrozek-Algebraic-Topology-A-Computational-Approach

09/30/2011

text

original

T. Kaczynski Universite de Sherbrooke K. Mischaikow Georgia Institute of Technology M. Mrozek Jagellonian University April 18, 2000

2

Contents

1 Introduction

1.1 Basic Notions from Topology . 1.1.1 Topological Spaces . . 1.1.2 Continuous Maps . . . 1.1.3 Connectedness . . . . . 1.2 Linear Algebra . . . . . . . . 1.2.1 Fields . . . . . . . . . 1.2.2 Vector Spaces . . . . . 1.2.3 Linear Maps . . . . . . 1.2.4 Quotient Spaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 9 16 22 24 25 27 32 33

7

2 Motivating Examples

2.1 Topology . . . . . . . . . . . . . . . . . . . . 2.1.1 Homotopy . . . . . . . . . . . . . . . 2.1.2 Graphs . . . . . . . . . . . . . . . . . 2.1.3 A Preview of Homology . . . . . . . 2.1.4 Z2 Homology of Graphs . . . . . . . 2.2 Approximation of Maps . . . . . . . . . . . 2.2.1 Approximating Maps on an Interval . 2.2.2 Constructing Chain Maps . . . . . . 2.2.3 Maps of the Circle . . . . . . . . . . 3.1 3.2 3.3 3.4 3.5 Groups . . . . . . . . . . . . . . . . . . . Products and Sums . . . . . . . . . . . . Quotients . . . . . . . . . . . . . . . . . Homomorphisms . . . . . . . . . . . . . Matrix Algebra over Z and Normal Form 3 . . . . . . . . . .

37

37 38 41 43 50 56 56 62 68 77 83 87 89 94

3 Abelian Groups

77

4

CONTENTS

3.6 Decomposition Theorem for Abelian Groups . . . . . . . . . . 101 3.7 Homology Groups . . . . . . . . . . . . . . . . . . . . . . . . . 105

4 Cubical Homology

4.1 Cubical Sets . . . . . . . . . . . . . . . . . . . . . . . . . 4.1.1 Elementary Cubes . . . . . . . . . . . . . . . . . 4.1.2 Representable Sets . . . . . . . . . . . . . . . . . 4.1.3 Cubical Sets . . . . . . . . . . . . . . . . . . . . . 4.2 The Algebra of Cubical Sets . . . . . . . . . . . . . . . . 4.2.1 Cubical Chains . . . . . . . . . . . . . . . . . . . 4.2.2 The Boundary Operator . . . . . . . . . . . . . . 4.2.3 Homology of Cubical Sets . . . . . . . . . . . . . 4.3 H0 (X ) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4 Elementary Collapses . . . . . . . . . . . . . . . . . . . . 4.5 Acyclic Cubical Spaces . . . . . . . . . . . . . . . . . . . 4.6 Reduced Homology . . . . . . . . . . . . . . . . . . . . . 4.7 Comparison with Simplicial Homology . . . . . . . . . . 4.7.1 Simplexes and triangulations . . . . . . . . . . . . 4.7.2 Simplicial Homology . . . . . . . . . . . . . . . . 4.7.3 Comparison of Cubical and Simplicial Homology . 5.1 5.2 5.3 5.4 Chain Maps . . . . . . . . . . . . . . Cubical Multivalued maps. . . . . . . Chain Selectors. . . . . . . . . . . . . Homology of continuous maps. . . . . 5.4.1 Cubical Approximations . . . 5.4.2 Rescaling . . . . . . . . . . . 5.4.3 Homotopy Invariance of Maps 5.5 Lefschetz Fixed Point Theorem . . . 6.1 6.2 6.3 6.4 6.5 Relative Homology . . . . . . . . Exact Sequences . . . . . . . . . The Connecting Homomorphism . Relative Lefschetz Theorem . . . Mayer-Vietoris Sequence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. 110 . 110 . 113 . 118 . 122 . 122 . 128 . 132 . 137 . 140 . 147 . 155 . 158 . 158 . 161 . 165 . 171 . 176 . 180 . 184 . 185 . 189 . 196 . 199 . 206 . 211 . 214 . 215 . 215

109

5 Homology of Maps

171

6 Homological Algebra

205

CONTENTS 5 A Equivalence Relations 217 .

6 CONTENTS .

The focus was on di erentiable functions and how they can be best approximated locally by linear functions (the derivative). then you began by learning about functions of the real line. i. However. In particular. 2)? Trying to nd an explicit inverse is di cult. xy 2y2 4x 1) invertible near the point (1. Along the way you learned about continuous functions. Di erenti7 . it allows one to study the local and global properties of continuous functions between general spaces. Again. linear functions can be studied algebraically as you learned in your linear algebra course. Topology incorporates further generalizations. a notable exception is the intermediate value theorem. of course. As was mentioned above one of the powers of the calculus is that through di erentiation di erentiable functions are locally approximated by linear functions. y) = (x2 3xy + y + 2. much easier to work with.e.Chapter 1 Introduction The subject of Topology grew out of the foundations of calculus and more generally analysis. the emphasis was on local properties such as limits. This introductory chapter summarizes the elementary topology which we will need. As an example of the advantage gained by this process of algebratization consider the following question. calculus gives us a simpler way to answer the question. Linear functions are. Is the function f : R2 ! R2 given by f (x. functions from Rn to Rm. If you took a typical calculus sequence. To read this book you do not need to have studied general topology. Furthermore. Later on these concepts were generalized to functions of more than one variable.

D 1](f2)(x. 'Df(1. this method allows us to develop an algorithmic approach to answering this question.y) -> array(1. then there is an open neighborhood V U containing x such that f : V ! f (V ) is invertible with a di erentiable inverse.y)]]): 'f(x.D 2](f1)(x. The important point of this example is that through calculus we have reduced an analytic problem to an algebraic problem. 1.2).y). For example using the computer package MAPLE we can solve this problem as follows. with(linalg): f1 := (x. the derivative of f at x. Let x 2 U .y) -> (f1(x.8 ating f gives CHAPTER 1. Of course this is just a special case of the following theorem. 'Det(Df(1. 2) = " 4 2 2 : 7 # Since the determinant of this matrix does not equal zero. INTRODUCTION " # 3 Df (x.y).D 2](f2)(x.2)).f2(x. In fact. 'Df(x.1 Inverse Function Theorem] Let U be an open set in Rn and let f : U ! Rn be a di erentiable function. On a super cial level we might say that calculus.2*y^2 -4*x-1: f := (x. is an invertible matrix. If Df (x). through the derivative.y)'=Df(x.y) -> x^2 -3*x*y +y +2: f2 := (x. Theorem 1. f is invertible in a neighborhood of (1. y) = 2x 4y x3x +y1 : y 4 Evaluating this at (1.2)'=Df(1.y)'=f(x. D 1](f1)(x.y) -> x*y .y).y).y)): Df := (x.2))'=det(Df(1.2.2.y)]. 2).y). provides us with a way to transform the study of local properties of ... 2) we get Df (1.

Any nite intersection of elements of T is in T . BASIC NOTIONS FROM TOPOLOGY 9 di erentiable functions to problems in linear algebra. 1.it will be introduced when the time comes). and X are in T . familiarity with the basic ideas of topology is worthwhile for at least two reasons. De nition 1. as in the case of all important mathematics.1 Basic Notions from Topology It was stated above that knowledge of general topology is not a prerequisite for this book. There are several di erent algebraic structures that can be assigned to topological spaces. First. . or more precisely group theory (don't worry about what a group is at this moment . 2.1. 3. As will be shown in this book algebraic topology provides a means by which one can transform the study of the global properties of topological spaces and continuous functions to problems in algebra. The elements of the topology T are called open sets. 1. it is hoped that after nishing this book you will be motivated to continue your study of topology. the one we will study is called homology. and therefore. you may as well begin using the language of topology at this point. Our focus will be on developing an algorithmic approach to homology theory which allows us to use the computer to solve topological problems.1. many problems can be reduced to simple symbolic computations as described above. the abstraction helps clarify the essential ideas.2 A topology on a set X is a collection T of subsets of X with . the following properties: 1. A set X for which a topology T has been speci ed is called a topological space. since for elementary functions many of the operations used in calculus can be implemented as algorithms and since linear algebra is also amenable to algorithmic implementation. Second.1. Furthermore. Any union of elements of T is in T .1 Topological Spaces The most fundamental de nition is that of a topological space. While this is correct.

To prove this let x 2 ( 1. De nition 1. another way of saying this is that the open sets in Rn can be de ned in terms of the Euclidean norm. (a.1). This topology is called the standard topology on Rn. both r0 > 0 and r1 > 0.5 Any interval of the form (a. Let = minfr0. b) is open in R.10 CHAPTER 1. Example 1. . 2) is an open set in the standard topology on R. : : : . The Euclidean norm of x is given by jjxjj2 := x2 + x2 + 1 2 q + x2 : n Given a point x 2 Rn. Since this is true for any x 2 ( 1. Generalizing this argument leads to the following result. This is equivalent to the conditions 1 < x and x < 2.fortunately we don't need to work at this level of generality. INTRODUCTION This is a fairly abstract de nition . Let x = (x1 . Then.3 A set U Rn is open if and only if for every point x 2 U there exists an > 0 such that B2 (x. 2). Since a topology is nothing but a collection of sets that satis es the conditions of De nition 1. r) := fy 2 Rn j jjx yjj2 < rg: The topology on Rn is typically de ned in terms of the Euclidean norm. Choose r0 = (x + 1)=2 and r1 = (2 x)=2.4 The interval ( 1. B2(x. r1 g. Proposition 1. xn) 2 Rn. Thus. In fact in everything we do we will always assume that the set X is a subset of Rn and that X inherits the standard topology from Rn. 1) or ( 1. 2). To explain what we mean by this recall the following ideas from analytic geometry. 2) R is an open set in the standard topology on R. Unless it is explicitly stated otherwise Rn will always be chosen to be the topological space speci ed by the standard topology. ) ( 1. we have shown that ( 1.2. the ball of radius r > 0 centered at x is given by B2 (x. 2). b). ) U . The reader should check that this de nition of an open set is consistent with the de nition of a topology (see Exercise 1.

1] is not open in the standard topology on R. b) or a. 1] R. (a. b) (c.6 The unit n-ball . Since open sets play such an important role in topology it is useful to be able to refer to the largest open set contained by a set. it follows that the arbitrary union of intervals is open. In fact. b] is not open in the standard topology on R. As an example consider (0. 1 i n Example 1. Therefore. De nition 1.1. 1]. r) Dn. 1]. Observe that if x 2 Dn then jjxjj2 < 1. For our purposes the supremum norm which is de ned by jjxjj := sup jxij for x = (x1 . Given a point x 2 Rn.2. Then. ) 6 (0.1) As before we can use this norm to de ne a collection of sets. Example 1. From De nition 1.g.7 Of course not every set is open. 1 2 (0. One of the advantages of the abstract de nition of a topology is that it does not explicitly involve a particular norm or distance. The interior of A is denoted by int (A) Since the arbitrary union of open sets is open.8 The interior of a set A is the union of all open sets contained in A. 0 < 1 jjxjj2. ) := fy 2 Rn j jjx yjj < g: Since the supremum norm represents a di erent way of measuring distance an -cube is di erent from an -ball (see Figure 1. The same argument shows that any interval of the form (a.1. e. but given any > 0. the -cube centered at x is B (x. int (A) is an open set. : : : .3. BASIC NOTIONS FROM TOPOLOGY 11 Proof: See Exercise 1. there are other norms that can be put on Rn which give rise to the same topology. B2 (x. d) is an open set. (0. Dn := fx 2 Rn j jjxjj2 < 1g is an open set in the standard topology on Rn. Therefore. xn) 2 Rn is particularly convenient. b]. a. B2(1.2. Let r = 1 jj2xjj2 .

). INTRODUCTION t(0.3 which means that V is in the standard topology. As important as an open set is the notion of a closed set. X n K := fx 2 X j x 62 K g . De nition 1. 1) t(1. ) V . the reader should check that Tsup de nes a topology on Rn (see Exercise 1. ) B (x. Then there exists > 0 such that B2(x. 0) &% fx 2 R2 j jjxjj2 = 1g fx 2 R2 j jjxjj = 1g Figure 1. One can check that B (x. 0) t(0.10 Tsup is the same as the standard topology on Rn.12 CHAPTER 1. Proof: To prove this result it needs to be shown that every set V 2 Tsup is in the standard topology and every set in the standard topology is in Tsup. Let U be an open set in the standard topology. Observe that B2 (x. ). ) V . Let x 2 V . ) U . V satis es De nition 1. 1) '$ t(1. Proposition 1.2). De nition 1. Let x 2 U . Then there exists > 0 such that B (x. pn ) B2(x. Therefore.1: The unit distance from the origin in the Euclidean norm and the unit distance from the origin in the Supremum norm. Thus U 2 Tsup . Again. Let V 2 Tsup .11 A subset K of a topological space X is closed if its complement is open.9 Let V 2 Tsup if and only if for every point x 2 V there exists > 0 such that B (x.

Similarly. Example 1. that fx 2 Rn j jjxjj > 1g is open.e. The closure of A in X is the intersection of all closed sets in X containing A. (3) See Exercise 1. This is equivalent to claiming that Rn n C n is open.) . Then the following statements are true. a. b] R are closed. BASIC NOTIONS FROM TOPOLOGY 13 Example 1. 1. Many sets are neither. X n T 2A K is open. Then B (x.16 Let X be a topological space and let A X . b] = ( 1. The closure of A is denoted by cl A (many authors also use the notation A. there exists at least one coordinate. the interval (0.14 The reader should take care not to get lulled into the idea that a set is either open or closed. As was observed in Example 1. a) (b.1. such that jxj j > 1..1. For example. say the j -th coordinate. it is not closed since its complement is ( 1. De nition 1. and X are closed sets. i. Finite unions of closed sets are closed. Therefore. 1) is open. by de nition X n K is open for each 2 A and the arbitrary union of open sets is open.8.7.15 Let X be a topological space. 0] (1. 2. (2) Let fK g 2A be an arbitrary collection of closed sets. jxj j 1 ) Rn n C n: 2 Remark 1. Proof: (1) . it is not open.13 The set C n := fx 2 Rn j jjxjj 1g is closed. . 1) which is not open. 1) and ( 1. T 2A K is closed. b] is a closed subset of R.:::. Thus. Similarly. Observe that jjxjj > 1 is equivalent to maxi=1. Arbitrary intersections of closed sets are closed. This is straightforward to see since its complement R n a. 1] R is neither open nor closed.12 The interval a.nfjxijg > 1. Then Xn \ 2A K = 2A (X n K ): Since. 3. Theorem 1. = X n X and X = X n .

1) = cl (0.19 Consider 0. therefore the closure of an arbitrary set is a closed set. di cult to prove. observe that A cl A and therefore cl A is the smallest closed set which contains A. So consider X Rn. The subspace topology on X is the collection of sets TX := fX \ U j U 2 T g: . 1] is closed by showing that ( 1. 1]: Example 1. 1) is not open. and u \ (X n A) 6= . but we begin with a more general de nition. 1)) = ( 1.14 CHAPTER 1. The abstract de nition of a topology only requires that one begin with a set X . 1) = 0. Therefore.17. 0] 1.. 1] = 0. Is there a natural way to specify a topology for X in such a way that it matches as closely as possible the topology on Rn? The answer is yes. Also. Finally one observes that any closed set that contains 0. U \ A 6= . 1]. 0) (1. 0) (1.17 Consider 0. 1). A point x 2 bd A if and only if for every open set U X containing x. Similar argument shows that cl (0. INTRODUCTION By Theorem 1. 1] R. 1) R. First one needs to check that 0. The boundary of A is de ned to be bd A := cl A \ cl (X n A): Example 1. cl 0. 1]. This follows from the fact that 1. 1]. 1) = cl 0. 1) must contain 0. This is not too De nition 1. Observe that cl (( 1. Proof: Up to this point. bd 0. Then one shows that 0. Then cl 0. Proposition 1. De nition 1. the only topological spaces that have been considered are those of Rn for di erent values of n. 1] = 0. Let X Z . From Example 1. 1] = cl 0. 1) is not closed.21 Let Z be a topological space with topology T .20 Let A X .18 Let X be a topological space and let A X . 1) is an open set in R.15 the arbitrary intersection of closed sets is closed. 1] = f0g f1g The following proposition gives a nice characterization of points that lie in the boundary of a set.

The intersection and union properties follow from the following equalities: n \ ! (X \ Ui ) = X \ Ui i=1 i=1 ! (X \ Ui ) = X \ Ui n \ i2I i2I for any indexing set I . 1]. Proof: The three conditions of De nition 1.1. .22 TX de nes a topology on X . X 2 TX since X = X \ Z. Observe that 1. 1] = (0. BASIC NOTIONS FROM TOPOLOGY 15 Before this de nition can be accepted. the sets themselves may \look" di erent. 2) \ 1. a) and (a. However.23 Consider the interval 1. thus both are open sets.3 de nes a topology for R2. 1]. 1] R with the subspace topology induced by the standard topology on R. Exercises 1. 1] = (0. 2) is an open set in R.. 1] is an open set in 1. 0) = X n (0. observe that . any set X Rn can be treated as a topological space. Example 1. 1. (0. 0) (0. hence Example 1. 0) = ( 2. We leave it to the reader to check that any interval of the form 1. 1] where 1 < a < 1 is an open set in 1.1.24 Let X = 1. (0. the following proposition needs to be proved. 2) \ X . 1] and (0. 1]. Using this de nition of the subspace topology.2 need to be checked. It is important to notice that while open sets in the subspace topology are de ned in terms of open sets in the ambient space. First. 0) so both are also closed sets. Proposition 1. Similarly. 2 TX since . 1] = X n 1.1 Prove that De nition 1. 0) \ X and (0. = X \ . This shows that for general topological spaces one can have nontrivial sets that are both open and closed.

16

CHAPTER 1. INTRODUCTION

1.2 Prove that Tsup de nes a topology for R2. 1.3 Prove Proposition 1.5. 1.4 Prove that any set consisting of a single point is closed in Rn. 1.5 Prove that B (x; pn ) B2 (x; ). 1.6 Let

Let Prove the following: 1. Qn 2.

n

Qn := fx 2 Rn j 0 xi 1g Rn:

n 1

:= bd Qn :

Rn is closed. 1 = fx 2 C n j x 2 f0; 1g for some i = 1; : : : ; ng. i

1.7 Let Z be a topological space with topology T . Let Y X Z . Let TX be the subspace topology obtained from viewing X Z . Let TY be the subspace topology obtained from viewing Y Z . Let SY be the subspace topology obtained from viewing Y X where X has the topology TX . Prove that SY = TY . 1.8 Prove that the nite intersection of closed sets is closed.

i = 1; 2; 3. Prove that Q is a closed set.

1.9 Let Q = k1; k1 + 1] k2; k2 + 1] k3 ; k3 + 1] R3 where ki 2 Z for

With the notion of subspace topology we have at our disposal a multitude of di erent topological spaces, in particular we can topologize any subset of Rn. A natural question is which topological spaces are \equivalent" and which are \di erent." The quotation marks are included because these terms need to be de ned before an answer can be given.

1.1.2 Continuous Maps

1.1. BASIC NOTIONS FROM TOPOLOGY

17

di erent from the geometric point of view: the rst one is a polyhedron, the second one is not. However, we would like to think of them as being \equivalent" in a topological sense, since they can be transformed from one to the other and back by simply stretching or contracting the spaces. To be more precise, observe that any element of Y has the form y = (r cos ; r sin ) where 0 r 1 and 0 =2. De ne f : Y ! X by

Example 1.25 The square X := 0; 1] 0; 1] R2 and a portion of the closed unit disk Y := fx 2 R2 j jjxjj 1; x1 0; x2 0g R2 are clearly

r tan ) =4, f (r cos ; r sin ) := (r;cot ; r) if 0 =4 (r if =2. Observe that this map just expands Y by moving points out along the rays emanating from the origin. One can also write down a map g : X ! Y which shrinks X onto Y along the same rays (see Exercise 1.10).

You have already seen maps of the form of f in the previous example in your calculus class under the label of a continuous functions. Since we introduced the notion of topology on an abstract level, we need to de ne continuous functions in an equally abstract way. Recall that a topological space consists of two objects, the set X and the topology T . Therefore, to compare two di erent topological spaces one needs to make a comparison of both the elements of the sets - this is done using functions - and one needs to compare the open sets that make up the two topologies.

De nition 1.26 Let X and Y be topological spaces with topologies TX and TY , respectively. A function f : X ! Y is continuous if and only if for every open set V 2 TY its preimage under f is open in X , i.e. f 1(V ) 2 TX :

Even in this very general setting we can check that some maps are continuous.

Proposition 1.27 Let X and Y be topological spaces. (i) The identity map 1X : X ! X is continuous.

18

CHAPTER 1. INTRODUCTION

(ii) Let y0 2 Y . The constant map f : X ! Y given by f (x) = y0 is continuous. Proof: (i) Since 1X is the identity map, 1X1 (U ) = U for every set U X . Thus, if U is open, its preimage under 1X is open. (ii) Let V Y be an open set. If y0 2 V then f 1(V ) = X which is open. If y0 62 V , then f 1(V ) = ; which is also open.

**Proposition 1.28 If f : X ! Y and g : Y ! Z are continuous maps, then g f : X ! Z is continuous.
**

Proof: Let W be an open set in Z . To show that g f is continuous we need to show that (g f ) 1(W ) is an open set. However, (g f ) 1(W ) = g 1(f 1(W )). Since f is continuous, f 1(W ) is open and since g is continuous g 1(f 1(W )) is open. This de nition tells us how we will compare topological spaces. Therefore, to say that two topological spaces are equivalent it seems natural to require that both objects, the sets and the topologies, be equivalent. On the level of set theory the equivalence of sets is usually taken to be the existence of a bijection. To be more precise, let X and Y be sets. A function f : X ! Y is an injection if for any two points x; z 2 X , f (x) = f (z) implies that x = z. f is a surjection if for any y 2 Y there exists x 2 X such that f (x) = y. If f is both an injection and a surjection then it is a bijection. If f is a bijection then one can de ne an inverse map f 1 : Y ! X .

De nition 1.29 Let X and Y be topological spaces with topologies TX and TY , respectively. A bijection f : X ! Y is a homeomorphism if and only if

both f and f

1

are continuous.

logical spaces.

Proposition 1.30 Homeomorphism de nes an equivalence relation on topo-

Proof: Recall (see A.2) that to show that homeomorphism de nes an equivalence relation we need to show that it is re exive, symmetric and transitive. To see that it is re exive, observe that given any topological space X the identity map 1X : X ! X is a homeomorphism from X to X . Assume that X is homeomorphic to Y . By de nition this implies that there exists a homeomophism f : X ! Y . Observe that f 1 : Y ! X is also

1.1. BASIC NOTIONS FROM TOPOLOGY

19

a homeomorphism and hence Y is homeomorphic to X . Thus, homeomorphism is a symmetric relation. Finally, Proposition 1.28 shows that homeomorphism is a transitive relation, that is if X is homeomorphic to Y and Y is homeomorphic to Z , then X is homeomorphic to Z . As before, we have introduced the notion of continuous function on a level of generality much greater than we need. The following result indicates that this abstract de nition matches that learned in calculus.

Theorem 1.31 Let f : R ! R. Then, f is continuous if and only if for every x 2 R and any > 0, there exists a > 0 such that if jx yj < then jf (x) f (y)j < .

Proof: ()) Let f : R ! R be continuous. Consider x 2 R and > 0. Observe that the interval B (f (x); ) = (f (x) ; f (x) + ) is an open set in the range of f . Since f is continuous, f 1(B (f (x); )) is an open set in R. Obviously x 2 f 1(B (f (x); )). Hence, by the de nition of an open set in the standard topology on R, there exists > 0 such that

B(x; ) = (x

; x + ) f 1(B (f (x); )):

We will now check that this is the desired . If y 2 R such that jx yj < , then y 2 (x ; x + ) and hence f (y) 2 B (f (x); ). Therefore, jf (x) f (y)j < . (() This direction is a bit more di cult since we have to check that for every open set V R, f 1(V ) is open. With this in mind, let V be an arbitrary open set in R. By de nition for each z 2 V there exists z > 0 such that B (z; z ) V . Observe that

V=

z2V

B (z; z ):

Assume for the moment that we can prove that for every z 2 V , f 1(B (z; z )) is open. Then we are done, since

f 1 (V ) =

z2V

f 1(B (z; z ))

and the arbitrary union of open sets is open.

z ): This implies that B (w.. 1 Let = 2 minff (w) z + z . a) for any a 2 R. Then there exists w 2 f 1(B (z. but su ciently small. B (f (w). then there exists > 0 such that jw yj < implies jf (x) f (y)j < . all that we need to prove is that given z 2 V and z > 0.32 Let f : Rn ! Rm . z )). f is continuous if and only if for every x 2 Rn and any > 0. f 1(x) = ln(x a) is also continuous. z )) is open. Proposition 1. With this in mind observe that it is possible that f 1(B (z. there exists a > 0 such that if jjx y jj < then jjf (x) f (y )jj < .33 The following topological spaces are homeomorphic: (i) R. Since w was an arbitrary element of f 1(B (z.. Thus. 1) for any a 2 R. z )) is open. then f 1(B (z. ) B (z. using Theorem 1. 1) be de ned by f (x) = a + ex: This is clearly continuous. z )). INTRODUCTION Thus. Proof: We begin by proving that R and (a. So assume that f 1(B (z. z ) We are nally ready to use the de nition of continuity from calculus. b) for any 1 < a < b < 1. z + z ). Furthermore. A straightforward generalization of this proof gives the following theorem Theorem 1. z ) = (z z . . This implies that f (w) 2 B (z. z )). Let = . z + z f (w)g. 1) are homeomorphic. is an open set.20 CHAPTER 1. Then. (iii) ( 1. (iv) (a. ) f 1(B (z. Another way of saying this is that f (B (w. )) B (f (w). This is okay since . z )) = . z )) 6= . f 1(B (z. Then. (ii) (a. Let f : R ! (a. ) B (z.31 we can easily show that a variety of simple topological spaces are homeomorphic.

. Therefore. 1].11 Prove Proposition 1. Now assume x 2 f 1(K ). But X n f 1(Y n U ) = f 1(U ). X n f 1(Y n K ) f 1(X ).10 Refering to Example 1. 1) ! ( 1. Proof: See Exercise 1. Since f is continuous. f (x) 2 K or equivalently x 2 f 1(K ). By hypothesis. Proof: ()) Let K Y be an a closed set. b) ! R given by a f (x) = ln x x b 1. Another useful way to characterize continuous functions is as follows. (b) Prove that f is a continuous function.35 Let f : X ! Y . 2. Let x 2 X n f 1(Y n K ). f is continuous if and only if for every closed set K Y . Thus. x 62 f 1(Y n K ) and hence x 2 X n f 1(Y n K ). Thus X n f 1(Y n U ) is open. BASIC NOTIONS FROM TOPOLOGY 21 is continuous and has a continuous inverse given by f 1(x) = (bey + a)=(1 + ey ). Hence X n f 1 (Y n X ) is closed in X .34. 1. Then f (x) 2 Y and f (x) 62 Y n K .1. a. f 1(Y n U ) is closed.34 The following topological spaces are homeomorphic: 1. (() Let U Y be an open set. b) is homeomorphic to R observe that f : (a. Proposition 1. Then Y n K is an open set. 1. 1) and hence to R. Observe that f : (a. f 1(Y n K ) is an open subset of X .11. b] for any 1 < a < b < 1.1. any iterval of the form ( 1. a) given by f (x) = x is a homeomorphism. Finally. f 1 (K ) is a closed subset of X . Exercises Proposition 1. Then.25: (a) Write down the inverse function for f . Thus. to see that (a. Then Y n U is a closed subset. it only needs to be shown that X n f 1(Y n K ) = f 1(K ). b) is homeomorphic to ( b. Thus.

Then X is a disconnected 1.24 the subspace topology. b] such .12 or the reader can consult 2]). If X is connected. where A 6= . Theorem 1.37 Let X = 1. and X . 1] are both open and closed in While it is easy to produce examples of disconnected spaces proving that a space is connected is more di cult. 0) (0.. Hints as to how to prove this theorem can be found in Exercise 1. 0) and (0.38 Any interval in R is connected. 1] space since by Example 1. that is both open and closed. Then there exists an set A Z . R. INTRODUCTION One of the most fundamental global properties of a topological space is whether or not it can be broken into two distinct open subsets. Proof: Let Z = f (X ).. f 1(A) is both open and closed.36 Let X be a topological space. But f 1(A) 6= . Example 1. We can now prove one of the more fundamental theorems from topology that you made use of in your calculus class. then there exists c 2 a.22 CHAPTER 1. then so is f (X ) Y . b] ! R is a continuous function and if f (a) > 0 and f (b) < 0.40 Intermediate Value Theorem] If f : a. that f (c) = 0. Z . The following de nition makes this precise.1. X is connected if the only subsets of X that are both open and closed are . If X is not connected then it is disconnected. Suppose that Z is disconnected. Theorem 1. Since f is continuous. Even the following intuitively obvious result is fairly di cult to prove.39 Let f : X ! Y be a continuous function.3 Connectedness De nition 1. 1. A very useful theorem is the following. X which contradicts the assumption that X is connected. Theorem 1.

1]. then either a < b or a > b.39. 1] ! 1. b]. 1] is a connected set. (0. contradicting Theorem 1. Use the fact that c is the least upper bound for A0 and that B0 is open. Exercises Example 1. b] such that f (c) = 0. 1) onto the set (0. Let c be the least upper bound for A0 . Then the restriction of f to (0. 1). We will argue by contradiction. c := inf fx 2 R j x > y for all y 2 A0 g: (d) Show c 62 B0. 1) is a homeomorphism and let t := f (1). With this in mind.1. 1): Let U = ( 1.12 This exercise leads to a proof that the interval 0. U and V are not trivial. 1) \ f ( a. 0) (0.39. b]) ( 1. Assume without loss of generality that a < b.1.41 The half-closed interval (0. b]) and V = (0. b] under the subspace topology. Let A0 := A \ a. Therefore. b]) is disconnected contradicting Theorems 1. 1] is not homeomorphic to the open interval (0. i. The following arguments will show that I 6= A B . 0) \ f ( a. b]). 1) is a homeomorphism of (0. (a) Show that the interval a. b] I . let A and B be two disjoint nonempty open sets in I = 0. b] and B0 := B \ a. Assume that there is no c 2 a. U and V are open sets and f ( a. b]) = U V . That is impossible since the rst set is connected and the second is not.38 and 1. 1).e. f ( a. (c) Show that c 2 a. Suppose that f : (0. Then f ( a. BASIC NOTIONS FROM TOPOLOGY 23 Proof: The proof is by contradiction. Using the subspace topology. . Let a 2 A and b 2 B . Since f (a) > 0 and f (b) < 0. b]. t) (t. (b) Show that A0 and B0 are open in a.

e. We shall review them. Prove that any simple closed 1.24 CHAPTER 1. Hint: Use an argument similar to that in Example 1. 1. b) Show that any closed bounded interval a. a point x 2 X such that f (x) = x. b] has the xed point property. 1.15 We say that a topological space X has the xed point property if every continuous map f : X ! X has a xed point. .13 Let A and B be connected sets. Finally. that it is invariant under a homeomorphism. i. b] under a continuous map : a. we do assume that you are familiar with the most basic ideas from linear algebra. Assume that A \ B 6= . that I 6= A0 B0. a) Show that the xed point property is a topological property. If the words feel unfamiliar don't worry they will be repeated many times throughout this text. t 2 a. INTRODUCTION (e) Show that c 62 A0 .e. but as in the previous section we shall present these ideas in a fairly general framework.2 Linear Algebra Homology theory (what we will learn in this book) provides an excellent geometric way to proceed from linear algebra to more abstract algebraic structures.17 A simple closed curve in Rn is an image of an interval a.14 Show that S 1 is connected. phic to an interval (whether it is closed. but c 62 A0 B0 and therefore.. 1. open or neither).16 Show that the unit circle S 1 = fx 2 R2 j kxk = 1g is not homeomor- curve is homeomorphic to a unit circle. 1.41. Again use the fact that c is the least upper bound for A0 and that A0 is open. Prove that 1. i. As was indicated earlier. Hint: Apply the Intermediate Values Theorem to the function f (x) x. b] if and only if s = a and t = b. A B is connected. b] ! Rn (called a path) such that (s) = (t) for any s < t. observe that c 2 I . s.

1 Fields x+y =y+x for all x.2. Let's review its properties in this context. 1. 1 is the identity element for multiplication. Addition is associative.1. 6. y 2 R. 4. LINEAR ALGEBRA 25 Let us begin with the real numbers R. x is the additive inverse of the element x. y. z 2 R. We usually write the operations as x+y and x y or simply xy. x (y z) = (x y) z for all x. 5. 2. Recall that there are two operations addition + : R R ! R and multiplication : R R ! R de ned on R. x y=y x for all x. x + (y + z) = (x + y) + z for all x. y. 0 is the identity element for addition. For each x 2 R there exists a unique element x 2 R such that x + ( x) = 0. . 7. z 2 R. Addition is commutative. Multiplication is commutative. In this section we will consider it to be a purely algebraic object. 3. In the previous section we were concerned with R as a topological space. 1.2. Multiplication is associative. There is a unique element 1 (one) in R such that x 1 = x for all x 2 R. There is a unique element 0 (zero) in R such that x + 0 = x for all x 2 R. y 2 R. The operations satisfy the following conditions.

that is x (y + z ) = x y + x z for all x. Example 1. Example 1. that satisfy properties (1) (9). the set of integers module 3.45 A very useful eld is Z2 . the set of integers module 2. y. These properties can be abstracted which leads to the notion of a eld. but 2 1 = 1 62 Z.44 The integers Z do not form a eld.43 The set of complex numbers C and the set of rational numbers Q are elds. x 1 is the multiplicative inverse of the element x. z 2 R. 2 2 Z. In particular. for addition and multiplication are as follows: + 0 1 2 0 0 0 1 2 0 0 1 1 2 0 1 0 2 2 0 1 2 0 Example 1. 2 Example 1.26 CHAPTER 1. Typically we simplify the expression of multiplication and write xy instead of x y. The rules for addition and multiplication are as follows: + 0 1 0 1 0 0 1 0 0 0 1 1 0 1 0 1 We leave it to the reader to check that properties (1)-(9) of a eld are satis ed. addition + : F F ! F and multiplication : F F ! F . INTRODUCTION 8. The rules 1 0 1 2 2 0 2 1 .42 A eld is a set F along with two operations. 9.46 Another eld is Z3 . De nition 1. For each x 2 R n f0g there exists a unique element x 1 2 R such that x x 1 = 1. Multiplication distributes over addition.

1. 1.2 Vector Spaces De nition 1. 1.18 Prove that the set of rational numbers Q is a eld. v 2 V then u + v 2 V and given 2 F and v 2 V . vector addition + : V V ! V and scalar multiplication F V ! V .2.48 A vector space over a eld F is a set V with two operations. Exercises 1. For which n is Zn a eld? In your linear algebra course you learned about vector spaces. we note that 1 = 2 and 2 1 = 2. As before let us think about this in an abstract manner. . Vector addition satis es the following conditions. However.47 Z4 .1. v+u= u+v for all vectors u.2. v 2 V . v 2 V . Vector addition is commutative. The rst time through you should read the following de nition substituting R for the eld F and Rn for the vector space V . the set of integers module 4 is not a eld. if u. The rules for addition and multiplication are as follows: + 0 1 2 3 0 0 1 2 3 1 1 2 3 0 2 2 3 0 1 3 3 0 1 2 0 1 2 3 0 0 0 0 0 1 0 1 2 3 2 0 2 0 2 3 0 3 2 1 Observe that the element 2 1 62 Z4 . LINEAR ALGEBRA 27 Again we leave it to the reader to check that properties (1)-(9) of a eld are satis ed. Furthermore. most probably the real vector spaces Rn. Example 1.19 Let Zn denote the set of integers modulo n.

INTRODUCTION u + (v + w) = (u + v) + w for all vectors u. however. an arbitrary vector is just a sum of these vectors. For all . y and z p directions. Finally. this is the way most calculus textbooks introduce vectors. 1 times v equals v where 1 2 F is the unique element one in the eld. For every v 2 V and . For every v 2 V . e. Vector addition is associative. e.1) . 2 F and every v 2 V ( + )v = v + v: De nition 1. ignoring the formality for a moment. CHAPTER 1. This de nition of a vector space may look formidable. w 2 V . v = i+ j+ k (1. 4. Typically to describe the vector space R3 one is told that the symbols i.49 Let V and W be vector spaces over a eld F .g.g. Of course. v. The scalar multiplication satis es the following rules: 1. For each vector v 2 V there exists a unique vector v 2 V such that v + ( v) = 0. if W V. and k represent basis vectors pointing in the x. 2i or 3 j. There exists a unique zero vector 0 2 V such that v + 0 = v for all v 2V. For every 2 F and all u. v 2 V . (u + v) = u + v: 4. They can be scaled by multiplying by a real number. 3. 2 F ( v) = ( )v 3. j. W is a subspace of V . 2. 1i = i and 0i = 0 is the zero vector.28 2.

2. j. 2 R.50 Let i. Example 1. LINEAR ALGEBRA 29 where . and k represent basis vectors for a vector space over the eld Z2 . In 3 . 2 3 2 3 2 3 2 3 1 1 0 6 1 7 i+k =6 0 7 j+k = 6 1 7 i+j+k =6 1 7 i+j=4 5 4 5 4 5 415 . If we choose to write v as a column vector. one lets R 2 3 2 3 2 3 1 0 6 0 7 j=6 1 7 k=6 0 7 i=4 5 4 5 405 0 0 1 and then (1. This vector space is denoted by Z3 and the typical vector has 2 the form v = i+ j+ k where .1) becomes 2 3 2 3 2 3 2 6 1 7+ 6 0 7+ 6 0 7=6 v= 4 0 5 4 1 5 4 0 5 4 0 0 1 3 7: 5 (1. An equivalent but di erent formalism is the use of column vectors.2) Depending on the context we will use both formalisms in this book. then we would have 2 3 2 3 2 3 2 3 6 1 7 + 6 0 7 + 6 0 7 = 6 7: v= 4 0 5 4 1 5 4 0 5 4 5 1 0 0 Since Z2 has only two elements we can actually write out all the vectors in the vector space Z3 .1. The advantage of the abstract de nition of a vector space is that it allows us to talk about many di erent types of vector spaces. . . 2 Z2 . Using both sets of notation they are: 2 2 3 2 3 2 3 2 3 0 1 0 6 0 7 i=6 0 7 j=6 1 7 k=6 0 7 0=4 5 405 4 5 4 5 0 0 0 1 0 1 1 1 Observe that in this vector space each vector is its own additive inverse.

The set S spans V if every element v 2 V can be written as a nite sum of multiples of elements in S . v = i+ j+ k where . Z3 is not a vector space. Similarly.52 Let V be a vector space.53 If V is a nite dimensional vector space. : : : . j. of V have the same number of elements.51 One can try to do the same construction over the integers. : : : ng F . Nevertheless. i. The importance of this last statement will become clear in Chapter 3. then any two bases . To make it clear why in the de nition of a vector space we insist that the scalars form a eld we need to recall some of the most fundamental ideas from linear algebra. Since Z is not a eld we will not. by de nition. : : : . properties 1-4 of scalar multiplication also hold. vng S and f 1. One of the most important results concerning nite dimensional vector spaces is that it has a dimension. V is a nitedimensional vector space if it has a nite basis. Example 1. A set of vectors S V is linearly independent if for any nite set of vectors fv1 . and k be basis elements. The zero vector is given by 0i + 0j + 0k: and v is given by i + ( )j + ( )k. Let i.30 CHAPTER 1. since Z is not a eld. This addition is clearly associative and commutative. 2 Z. INTRODUCTION other hand we can mimic what has been done before and de ne an algebraic object which we will denote by Z3 .e. On the De nition 1. then it makes perfectly good sense to talk about linear combinations of these elements. vng S the only solution + nvn = 0 is 1 = 2 = = n = 0. to the equation 1 v1 + 2 v2 + v = 1v1 + 2v2 + + n vn for some collection fv1. . A basis for V is a linearly independent set of vectors in V which spans V . Theorem 1. get a vector space.

If we return to Example 1.20 Let Z3 denote the three dimensional vector space over the eld Z3 .56 In the proof of Proposition 1. 3 .2. : : : . This means that there are distinct vectors v1. The previous remark may seem somewhat trivial and esoteric. Thus we can rewrite (1. linear independence is lost.51 then we can see that Proposition 1. Proposition 1. 1 Observe that w is not in the span of S since 2 1 62 Z. fact that F was a eld. Assume that by adjoining w to S . n. LINEAR ALGEBRA 31 This theorem allows us to make the following de ntion.55 need not hold in Z3.3) as w = 1( 1 v1 + 2 v2 + : : : + nvn) which contradicts the assumption that w is not in the subspace spanned by S. 4 0 5> and w = 4 0 5 : 0 0 2 . De nition 1. but as we shall soon see it has a profound e ect on the homology groups of topological spaces.55 we made crucial use of the 1.1. Suppose w is a vector in V which is not in the subspace spanned by S . 4 1 5 . vn S and nonzero scalars 1 . 3 Write down all the elements of Z3 .54 The dimension of a vector space is the number of elements in a basis. in the eld F such that (1. Proof: The proof is by contradiction. Let 82 3 2 3 2 39 2 3 > 1 <6 7 6 0 7 6 0 7> = 6 0 7: S = >4 0 5 . but S fwg is not a linearly independent set. Exercises Remark 1. 1 2 F .55 Let S be a linearly independent subset of a vector space V .3) 1 v1 + 2 v2 + : : : + n vn + w = 0: Since F is a eld. Then the set obtained by adjoining w to S is linearly independent. : : : . A very closely related result is the following.

60. A fundamental result from linear algebra is the following. A linear map or linear operator from V to W is a function L : V ! W such that L( v + u) = (Lv) + Lu for all u. De nition 1. . Then. 1. then ker L is a subspace of V and image L is a subspace of W . Exercises 1. De nition 1. The vector spaces V and W are said to be isomorphic if there exists an isomorphism L : V ! W .59 Let L : V ! W be a linear map.57 Let V and W be vector spaces over a eld F .32 CHAPTER 1.21.3 Linear Maps We now turn to a brief discussion of maps between vector spaces. Proof: See Exercise 1. Theorem 1.22 Let L : R2 ! R2 be given by L= 1 1 1 1 " # Compute ker L and image L.2. V and W are isomorphic if and only if dim V = dim W . L is an isomorphism if L is invertible. Draw them as subspaces of R2.60 If L : V ! W be a linear map. The kernel of L is ker L := fv 2 V j Lv = 0g and the image of L is image L := fw 2 W j Lv = w for some v 2 V g: Proposition 1. INTRODUCTION 1.58 Let V and W be nite dimensional vector spaces over a eld F .21 Prove Proposition 1. v 2 V and all scalars 2 F .

2. Vector addition is de ned by v] + u] := v + u] for all u. De nition 1. v 2 V and scalar multiplication is given by v] := v] for all 2 F.1. u if and only if u v since v u 2 W if and only if u v 2 W . v] := fu 2 V j u v 2 W g: Observe that if w 2 W .2. We will return to this type of construction over and over again. 3. v u and u x implies v x since v u 2 W and u x 2 W implies that v u + u x = v x 2 W . v 2. the notion of a quotient space is absolutely fundamental in algebraic topology.62 The quotient space V=W is the vector space over F consisting of the set of equivalence classes de ned above. v v for all v 2 V since v v = 0 2 W .61 de nes an equivalence relation on elements of V .e. then w 0 and hence w] = 0]. Proof: To prove that is an equivalence relation we need to verify the following three properties: 1. Let us set v u if and only if v u 2 W: Proposition 1. Consider V and W . i. Because these equivalence classes are so important we will give them a special notation. Given v 2 V let v] denote the equivalence class of v under this equivalence relation.4 Quotient Spaces As will become clear in the next chapter. vector spaces over a eld F . with W a subspace of V . LINEAR ALGEBRA 33 1. v 2 V: .

every element of V=W is in the image of . Thus. Finally. " . the second coordinate is ignored. How can we use quotient spaces to resolve this? Let 0 W := fw 2 V j w = w g: 2 " # " # " # a a c b We can now consider the vector space V=W whose elements are the equiv- Observe that W is a subspace of V and in the induced equivalence class alence classes. Consider the linear map : V ! V=W given by the matrix = 1 0].23).e. Then a typical element of V has the form " # v = v1 : v 2 Let us now assume that we don't care about the value of the second coordinate.34 CHAPTER 1. we can represent an element of V=W by a single number x. A little intuition as to what this represents may be in order. 1 v2 v2 i. We can still add vectors. multiply by scalars and all the rest but it seems a bit ine cient to carry around the second coordinate since we are ignoring it. This means that as far as we are concerned " # " # 1 = 1 2 5 since they agree in the rst coordinate and we don't care about the value of the second coordinate. Observe that is surjective. This vector space is a 1-dimensional vector space. i. since we can easily recover the corresponding equivalence class by considering the set of vectors " # x V.e. INTRODUCTION We leave it to the reader to check that this does indeed de ne a vector space (see Problem 1. Then # " # v1 = 1 0] v1 = v ]. Consider the vector space V = R2. i.e. v2 Of course the best way to compare two di erent vector spaces is through linear transformations from one to the other. notice that ker = W .

What is the dimension of R2=W ? .2. 1. LINEAR ALGEBRA 35 for this example the process of creating a quotient space is equivalent to the existence of a particular linear map. 1. Exercises In particular.23 Prove that V=W as de ned in De nition 1. As will be made clear in Chapter 3.62 is a vector space over F .1. this is not a coincidence. prove that vector addition and scalar multiplication are well de ned operations.24 Let W be the subspace of R2 spanned by the vector " # 1 : 2 Draw a picture indicating the equivalence classes in R2=W .

36 CHAPTER 1. INTRODUCTION .

**Chapter 2 Motivating Examples
**

Why study algebraic topology? This chapter contains a description of problems where algebraic topological methods have proven useful. These problems have their origins in topology (not surprising), computer graphics, dynamical systems, parallel computing, and numerics. Obviously for such a broad set of issues a single chapter cannot do any of the topics justice. They are included solely for the purpose of motivating the formidable algebraic machinery we are about to start developing. This chapter is meant to be enjoyed in the sense of an entertaining story. Don't sweat the details - try to get a feeling for the big picture. We will return to these topics throughout the rest of this book.

2.1 Topology

The importance in linear algebra of the dimension of a vector space is that any two nite dimensional vector spaces (over the same eld) of the same dimension are isomorphic. In other words from the point of view of linear algebra they are indistinguishable. Said yet another way, the set of nite dimensional vector space can be classi ed according to a single natural number. Algebraic topology is an attempt to do a similar thing, but in the context of topological spaces. Since topological spaces are more varied than vector spaces, the classi cation is done in terms of algebraic objects rather than the natural numbers. As pertains to this book the goal is as follows. Given a topological space X we want to de ne an algebraic object H (X ), called the 37

38

CHAPTER 2. MOTIVATING EXAMPLES

homology of X , which is a topologically invariant; that is, if X and Y are homeomorphic then H (X ) and H (Y ) are isomorphic. Notice that we did not claim that homology classi es spaces up to homeomorphism. It is not true that if two spaces have the same homology, then they are homeomorphic. Unfortunately, the classi cation problem in topology is too di cult for any purely algebraic classi cation. In fact, this problem is so di cult, that mathematicians have pretty much given up trying to classify arbitrary topological spaces up to homeomorphism. Instead they study the weaker equivalence relation known as homotopy type. Before giving the de nition let us consider a motivating example. We begin by recalling the intermediate value theorem which we proved earlier (Theorem 1.40).

2.1.1 Homotopy

**Theorem 2.1 If f : a; b] ! R is a continuous function and if f (a) > 0 and f (b) < 0, then there exists c 2 a; b] such that f (c) = 0.
**

This is a model topological theorem. The function is only assumed to be continous, global rather than local information is assumed, i.e. the values of the end points are given, and yet one is still able to draw a conclusion concerning the behavior of the function on its domain. Homology provides us with a variety of algebraic tools for determining if there exists a point c such that f (c) = 0. But this process of going from topology to algebra loses information. This should not be surprising. Think back to calculus where one uses the derivative to obtain a linear approximation of the di erentiable function. Many di erent functions can have the same derivative at a point. To get a better approximation one has to use Taylor polynomials. In fact only analytic functions can be approximated exactly by their derivatives. What families of spaces or maps will give us the same algebraic toplogical information? To answer this consider again the intermediate value theorem. The only important points are the endpoints so let f; g : a; b] ! R be di erent continuous functions with f (a) > 0 and g(a) > 0, and f (b) < 0 and g(b) < 0. Now consider the family of functions F : a; b] 0; 1] ! R de ned by F (x; s) = (1 s)f (x) + sg(x):

2.1. TOPOLOGY

39

Observe that F ( ; 0) = f ( ) and F ( ; 1) = g( ). For any xed value of s 2 0; 1] we have yet another function F ( ; s) : a; b] ! R. Observe that F (a; s) = (1 s)f (a) + sg(a) > 0 and F (b; s) = (1 s)f (b) + sg(b) < 0 Thus, we can apply the intermediate value theorem to F ( ; s) for any s 2 0; 1]. This family of functions is a special case of what is known as a homotopy. De nition 2.2 Let X and Y be topological spaces. Let f; g : X ! Y be continuous functions. f is homotopic to g if there exists a continuous map F : X 0; 1] ! Y such that F (x; 0) = f (x) and F (x; 1) = g(x) for each x 2 X . The map F is called a homotopy between f and g. f homotopic to g is denoted by f g. It is fairly straight forward to check that homotopy is an equivalence relation (see Excercise 2.1). How does this help us with the classi cation problem in topology? Since homotopy is an equivalence relation it can be used to de ne an equivalence between topological spaces. De nition 2.3 Two topological spaces X and Y are homotopic if there exist continuous functions f : X ! Y and g : Y ! X such that g f 1X and f g 1Y where 1X and 1Y denote the identity maps. X homotopic to Y is denoted by X Y . Example 2.4 Two topological spaces can appear to be quite di erent and still be homotopic. For example it is clear that Rn is not homeomorphic to the point f0g. On the other hand these two spaces are homotopic. To see this let f : Rn ! f0g be de ned by f (x) = 0 and let g : f0g ! Rn be de ned by g(0) = 0. Observe that f g = 1f0g and hence f g 1f0g . To show that g f 1Rn consider the function F : Rn 0; 1] ! Rn de ned by F (x; s) = (1 s)x: Clearly, F (x; 0) = x = 1Rn and F (x; 1) = 0.

40

**CHAPTER 2. MOTIVATING EXAMPLES
**

A special case of homotopy is that of a deformation retract.

**De nition 2.5 Let A
**

continuous map F : X

X . A deformation retraction of X onto A is a 0; 1] ! X such that F (x; 0) = x for x 2 X F (x; 1) 2 A for x 2 X F (a; s) = a for a 2 A:

If such an F exists then A is called a deformation retract of X . It is easy to check that if A is a deformation retract of X and B is a deformation retract of A, then B is a deformation retract of X . by F (x; s) = (1 s)x.

**Example 2.6 f0g is a deformation retract of 0; 1]. De ne F : 0; 1] ! f0g
**

Homology has the property that if two spaces are homotopic then their homologies are the same. On the other hand, there are spaces with the same homologies which are not homotopic. Thus, the algebraic invariants that we will develop in this book are extremely crude measurements of the topology of the space. Still there are interesting problems to which one can apply homology theory.

Example 2.7 Let

There is no deformation retraction of n to a point. We include this example at this point to try to indicate that this is a nontrivial problem. In particular, we encourage you to try to nd a proof of this fact. As motivation for the study of this subject we assure you that once you know homology theory, this example will become a triviality.

Exercises

n := fx 2 Rn+1 j jjxjj = 1g:

on X and Y prove that f

**2.1 Prove that homotopy is an equivalence relation. 2.2 Let f; g : X ! Y be continuous maps. Under the following assumptions
**

g.

: : : . De nition 2. if you prove the last case. . and 2. 41 2. called vertices. e2 .1. TOPOLOGY X = Y = 0. 2. E ) where V is a nite set whose elements are called vertices and E is a collection of pairs of distinct elements of V called edges. One of the important properties of homology is that it can be determined from combinatorial information. called edges which satisfy the following intersection conditions: 1. if an edge and a vertex intersect. k 1. : : : vng. ek such that ej \ ej+1 6= .1. for j = 1.2 Graphs De nition 2. To motivate the ideas and build some intuition before beginning with the formal de nitions it is useful to have a simple but large class of topological spaces. Up to now we have given no indication how one moves from the topology to the algebra.3 Prove that Rn n f0g is homotopic to S n 1. and ek \ e1 6= . together with straight line segments fe1 . then you have proven the rst two. A graph which is connected and has no loops is called a tree. A loop L in the graph is a union of edges e1. 1] X is any topological space and y 2 Y is a deformation retract of Y .2. : : : . 1] X = 1 and Y = 0. then they intersect at a unique vertex. With this in mind we present the following de nition which indicates there is a natural reduction of a nite graph to a combinatorial object. : : : .. emg. then the vertex is an endpoint of the edge.9 An abstract nite graph is a pair (V . joining vertices.8 A nite graph G consists of a nite collection of points in R3 fv1 . Obviously. if two edges intersect nontrivially.

0) = h(0 h 1(x)) = h(0) = v . Proof: Assume not. 1] ! e be such a homeomorphism with the property that h(0) = v and h(1) = v+. Then fej . MOTIVATING EXAMPLES v1 t e1 tv2 e5 v5 t @ v2 @t 3 e @ 2 e3 v v1 t e 3 e1 t @ @ @t 3 e2 v e4 t v4 tv4 Figure 2. s) = h(sh 1 (x)). ej+1. Fe de nes a retract of e to v . Observe that F (x. In particular.42 CHAPTER 2.1: A loop and a tree. Since T has no free vertices. there is an edge e2 with vertices + v2 such that v1 = v2 . Then there exists a tree T with 0 free vertices. De ne Fe : e 0. Fe(x. Therefore. at some point vi+ = vj for some i > j 1. . Since e is a line segment it is homeomorphic to 0. Proof: Let e be an edge with vertices v and v +. Let e1 be an edge in T . Proposition 2. Let n be the number of edges in T . Before turning to the algebra we want to consider the topology of trees. we will show that any tree is homotopically equivalent to a single point.11 Every edge is homotopic to a point. eig forms a loop. A vertex which only intersects a single edge is called a free vertex. 1) = h(h 1 (x)) = x and hence is the identity. Continuing in this manner we can label the edges by ei and the vertices by vi where vi = vi+ 1. Label its vertices by + v1 and v1 . 1]. Lemma 2. This is a contradition. : : : . Let h : 0.10 Every tree which contains at least one free vertex. Note since there are only a nite number of vertices. 1] ! e by Fe (x.

The result now follows by induction.1. In fact by Proposition 2.2.7 it was stated that 1 is not homotopic to a point. as the following example indicates di erent graphs can give rise to the same subset. 2. Let e be the edge which contains the vertex v+. Since a point is the simplest nontrivial topological space. We need to show it is true for a tree with n + 1 edges. we need to begin with a word of caution.10 T has a free vertex v+. observe that given our de nition. Let the other vertex of e be denoted by v .11 this is homotopic to a point. However. As such it will be dealt with much later. a nite graph is a xed subset of R3 .1. This shows that G is a deformation retraction of T onto T 0. Assume that the result is true for all trees with n edges or less.12 Every tree T contains a vertex v such that there exists a edges? 2. The proof that homology is a topological invariant is fairly complicated. s) = x (x. The homotopy is the deformation retraction.6 we showed that an interval is homotopic to a point. By Lemma 2. We will use this contrast to motivate how homology can be used to measure the di erence in the complexity of these two topological spaces.3 A Preview of Homology . In Example 2. Let T 0 be the tree obtained from T by removing the edge e and the vertex v+. s) if x 2 T Fe if x 2 e where Fe is de ned as in Lemma 2.11. 1] ! T by ( 0 G(x. Now de ne G : T 0.4 Up to homotopy how many di erent planar graphs are there with 5 In Example 2. By Proposition 2. Proof: The proof is by induction on the number of edges in the tree. However. TOPOLOGY 43 deformation retraction of T onto v . To make clear at the outset why working with a graph is not su cient to establish the topological invariance of homology.12 every tree is homotopic to a point. up to homotopy trees must be fairly simple topological spaces. Let T be a tree with n + 1 edges. The simplest tree consists of a single edge. Exercises Proposition 2.

1). MOTIVATING EXAMPLES topological space has many di erent representations as a graph. b] = fa. More precisely given points x 2 ( 1.13 Consider the family of graphs de ned by Gn = ffj=ng j j = 0. c. y 2 1. b be two distinct vertices in R2. The set of all extreme points of G is called the geometric boundary of G and denoted by bd G. We shall try to measure this distinction algebraically. d] and d.44 CHAPTER 2.14 A point x of a graph G is called a regular point of G if a su ciently small ball in G around x is homeomorphic to an open interval. 1] is represented by a graph consisting of four intervals a. Having made these explicit disclaimers we now take the liberty of using language in which we are implicitly assuming that we are working with a topological invariant. : : : ng f j=n. then there exists a homeomorphism between Ux and Vy . 0. (j + 1)=n] j j = 0. n 1g Observe that as a subset of R each graph describes 0. the points a. Example 2. ( 1. b. : : : . Ux and Vy of these points. Then bd 0. the same De nition 2. e]. 1] and 1 are topologically di erent. This is not trivial. It is worth making an observation at this point. 1) and 1 are indistinguishable. To be precise consider the graphs indicated in Figure 2. Thus to prove topological invariance we would have to show that given any two abstract graphs that are associated to nite graphs that in turn represent homeomorphic spaces the corresponding homology is the same. The notion of topological boundary is ambiguous here because it depends on the outer space the graph is imbedded to. A point which is not a regular point is called an extreme point of G. in the topology of a. b] = a. b] in the topolgy of R2. b] in the sense that they are extreme points of the interval. b are clearly distinct from the other points of a. Let us now think of 0. 1].2. 1] = . let a. b]. A word of caution is needed before we go further. b]. But no matter what is the outer space. Locally. and su ciently small neighborhoods. bg in the topology of the line passing through a and b. . That distinction is exhibited by the following de nition. and bd a. Locally the only di erence between 0. 1]. 1] and 1 as graphs. 1g of 0. For instance. Thus. With this in mind we begin by asking the question how can we show that 0. In our motivation of homology we will use abstract graphs. bd a. c]. 1] and 1 are the boundary points f0.

c]. c. What should we use for the scalars? A possible idea is Z2 . Recalling the discussion in the previous chapter where we described vector spaces. Topology bd a. For example fag becomes a. d. e] = d + e Table 2. 1]. fbg. a]g V 0 = ffag. d. b]. d] = fcg bd d. b] = a + b @ b. To keep our computations local we indicated in the left hand column of Table 2.1. d] = c + d @ d. c] = b + c @ c. On the topological level addition and subtraction of edges and points is not an obvious concept.this way. d]. however. b. On our ctional algebraic level. TOPOLOGY d t a t t c t b t t t t 45 b c d e E = f a. fcg. 1] are where we can see a di erence in local topology.2: Finite graphs and corresponding abstract nite graphs for 0. e g a t E 0 = f a. e] = fdg Algebra fbg fcg fdg feg @ a. c. 1] and 1 We mentioned earlier that the boundary points of 0. c. b. we write the algebraic objects in bold and allow ourselves to formally add them. if we make @ .1 the topological boundaries of each of the edges. d. b.2. b]. In the right hand column are what for the moment can be considered ctional algebraic quantities derived from the corresponding elements of the abstract nite graph. b] = fag bd b.1: Topological and algebric boundaries in 0. e]g V = fa. d]. fdg g Figure 2. c] = fbg bd c. c]. we will allow ourselves this luxury.

Adding up the algebraic boundaries we have @ ( a. algebraic objects whose boundaries add up to zero. Since there exists a deformation retract of C 2 to a point we need to understand . i. b] + b. d] = fcg bd d. are topologically nontrivial. a ] = d + a Table 2. d] = c + d @ d. d] + d. we can match the topological expression b. b] + b. observe that 1 C 2. b] a linear operator. c] + c.2: Topology and algebra of boundaries in 1.46 CHAPTER 2.1) Topology bd a. c]) = @ ( a. MOTIVATING EXAMPLES bd ( a. Based on these two examples one might make the extravagent claim that spaces with cycles. b] = fag bd b. 1] = fag feg. e]) = a + b + b + c + c + d + d + e = a + e: As an indication that we are not too far o track observe that on the topological level bd 0. b] + b. a] = fdg Algebra fbg fcg fdg fag @ a. c]) = a+b+b+c = a + 2b + c = a + c: Continuing in this way we have that @ ( a. b ] = a + b @ b. b]) + @ ( b.2. and in fact 1 = bd C 2. To see how this fails. c] = b + c @ c. c] + c. c] = fbg bd c. Doing the same for the graph and abstract graph representing 1 we get Table 2. c]) = fag fcg with the algebraic expression @ ( a. a]) = 0: (2.e. d] + d. This is almost true.

c] = fbg bd c. a]: Equation (2. d] + d. This is coded in the combinatorial information as the element fC 2g. c] c. b ] = a + b @ b. b. d] + d. a] @ C2 = a. a ] = d + a Table 2.2 contained in Table 2. c]. c] + c. a]g V 0 = fa. c] + c. b] bd a. c. d] + d.3. how the nontrivial algebra in 1 becomes trivialized. a] = fdg Algebra 47 b.1) indicated that the cycle a. a] was the interesting algebraic aspect of 1 . c. b].2. b] + b. b] + b. d] d. For the moment consider the picture and collection of sets in Figure 2.3: Simplicial complex and corresponding abstract simplicial complex for C 2 . d] = c + d @ d. The new aspect is the square C 2.1. d t a t tc tb fC 2 g E 0 = f a. Now observe that @ C2 = a. To do this we need to go beyond graphs into cubical complexes which will be de ned later. c] = b + c @ c. In C 2 it appears as the boundary of an . Since 1 C 2 . Table 2. b. one should expect to see the contents of Table 2.3 contains the topological boundary information and ctional algebra that we are associating to it for C 2. d] = fcg bd d. b] = fag bd b. d.3: Topology and algebra of boundaries in C 2. b] + b. a] fbg fcg fdg fag @ a. c] + c.3. d]. TOPOLOGY Topology bd C 2 = a. d g Figure 2.

MOTIVATING EXAMPLES object. b] + b.2 except that we have added arrows to our graphs to suggest a direction (the fancy word is orientation) through which we traverse the interval. So let us declare that @ ( a. i. we have indicated a direction through which we can traverse the loop 1. In fact. Are there other choices that make sense? Consider Figure 2. the image of this operator. c] + c. then we wish to ignore it. The process of developing new mathematics typically involves developing new intuitions and nding new patterns . For now lets just enjoy trying to match topology and algebra.a sum of algebraic objects whose boundaries add up to zero. i. Furthermore. d] + d.4 Again. elements of the kernel of some operator. . Furthermore. e]) = b a + c b + d c + e d = e a: Again we see that there is consistency between the algebra and the topology since bd 0. if this cycle is a boundary. 1] = feg fag and the arrows suggest traversing from a to e.48 CHAPTER 2. there are a lot of loose ends that we need to tie up and we will begin to do so in the next chapter. Recall that earlier we asked the question what should be use for scalars? We chose Z2 last time.don't worry. be happy! Admittedly. we see that the algebra that corresponds to the interesting topology is a cycle .in this case we have the advantage of knowing that it will all work out in the end. using the integers we can assign a plus or a minus sign to the edge or vertex depending on whether we traverse it following the assigned direction or not. Similarly.e.4 which looks alot like Figure 2. lets do it again. Doing the same for the graph and abstract graph representing 1 gives rise to Table 2. Restating this purely algebraically what we are looking for are cycles. Let us denote this operator by @ to remind us that it should be related to taking the boundary of a topological space. The observation that we will make is that cycles which are boundaries should be considered trivial. If you feel things are spinning out of control . In other words we are interest in an algebraic quantity which takes the form kernel of @=image of @: We have by now introduced many vague and complicated notions.e. We could argue that Z is a natural choice since it is not clear what a fractional amount of a vertex or an edge of an abstract graph should represent.

5. As before @ C2 = a. b] + b.5 contains the topological boundary information and ctional algebra that we are associating to it for C 2. d.4: Topology and algebra of boundaries in using Z coe cients. Again. c] + c.1). d]. fdg g Figure 2. fcg. the observation that we will make is: cycles which are boundaries should be considered trivial. c. a > = a 1 a b c d Table 2.2.1.4 contained in Table 2. d. d. b. b > = b @< b. d]. Consider Figure 2.4: Finite graphs and corresponding abstract nite graphs for 0. b. a]g V 0 = fag. d > = d @< d. b]. b]. Table 2. a] = fdg fbg fcg fdg fag @< a. b. TOPOLOGY d t a t - 49 t c 6 - b t - c t - d t - e t ? E = f a. d] + d. b] = fag bd b. More precisely we again arrive at equation (2. c].5. e]g V = fa. We still need to understand what happens to this algebra when we consider 1 C 2. Topology Algebra bd a. c. d] = fcg bd d. a] was the interesting algebraic aspect of 1 . . Since 1 C 2. c > = c @< c. e g a t t b E 0 = f a. In C 2 it appears as the boundary of an object. fbg. d] + d. 1] and 1 with a sense of direction. c]. c] + c. b] + b. c. a]: Equation (2.1) indicated that the cycle a. c] = fbg bd c. we again see the contents of Table 2.

We have done the same example twice using di erent scalars but the conclusion was the same. d g Figure 2. Topology Algebra bd C 2 = 1 = a.5: Topology and algebra of boundaries in C 2. c]. MOTIVATING EXAMPLES d t ? tc 6 6 - a t tb fC 2g E 0 = f a. b. d] + d.5: Simplicial complex and corresponding abstract simplicial complex for C 2.1. a little slower and more formally. b] b.5 Repeat the above computations for a graph which represents a triangle in the plane. c] + c. but in the general setting of graphs using the algebra of vector spaces. d] = d @ d.50 CHAPTER 2. c. d] d. This is still pretty fuzzy so lets do it again. c. Then. c] = c @ c. a]g V 0 = fa. b. b] + b. c] = fbg bd c. d]. Exercises 2. We should look for a linear operator that somehow algebraically mimics what is done by taking the topological boundary. a] = fdg fbg fcg fdg fag @ a. having found this operator we should look for cycles (elements of the kernel) but ignore boundaries (elements of the image). c] c. a] = a a b c d Table 2. a] bd a. a] @ C2 = a. d. 2.4 Z2 Homology of Graphs . b] = fag bd b. b]. d] = fcg bd d. b] = b @ b.

Let the edge ei have vertices vj and vk . The vector spaces Ci(G.e. Z2 ) takes the form v = 1 v1 + 2 v 2 + + nvn where i 2 Z2 . Z2 ) as follows. For graphs we will also set C2(G. Similarly. we can think of @0 . Z2 ) takes the form e = 1e1 + 2e2 + + k ek where i 2 Z2 . Z2) and C 1(G. Z2) be linear maps. Let G0 denote the set of vertices of G and let G1 denote the set of edges of G. Z2 ) ! C0(G.2. Z2) are called the i-chains for G. We will always take C 1(G. In line with the previous discussion we make the following de nition. Z2 ) and C1(G. Z2) @1 : C1(G. : : : . Since we have chosen bases for these vector spaces.1. @2 is the zero matrix. Since C 1(G. Z2) and again let the scalar eld be Z2 . The entries of the matrix @1 are determined by how @1 acts on the basis elements. Z2 ) and the typical element of C0(G. ng is a basis for C0(G. Z2 ) ! C 1(G. Thus.e. Z2) @2 : C2(G. Z2 ). kg is a basis for C1(G. TOPOLOGY 51 Let G be an abstract graph. We now need to formally de ne the boundary operators that were alluded to earlier. then the collection fei j i = 1. then the collection fvi j i = 1. It is convenient to introduce two more vector spaces C2(G. ek g. the vector space consisting of exactly one element 0. De ne @1 ei := vj + vk : . i. vng. Z2) and let the scalar eld be Z2 . We will construct two vector spaces C0(G. Z2 ) and the typical element of C1(G. : : : . let the set of edges G1 be the set of basis elements of C1 (G. @1 and @2 as matrices. If G1 = fe1 . : : : . Z2 ) ! C1(G. Z2) = 0. : : : . Declare the set of vertices G0 to be the set of basis elements of C0(G. if G0 = fv1 . i. the edges ei. it is clear that @0 must be the matrix with all zeros. Z2 ) to be the trivial vector space. As we will see later for more complicated spaces this need not be the case. Let @0 : C0(G. Similarly. Z2 ) to be the trivial vector space.

Let us compute the homology for the graphs in Figure 2. Z2) j @1 v = 0g Since @0 = 0 it is obvious that Z0(G. d. d]. b].e. To formally state this. i. Then. Z2 ) j 9 e 2 C1(G. MOTIVATING EXAMPLES In our earlier example we were interested in cycles. Z2 ). eg G1 = f a. elements of the kernel of the boundary operator. Z2) := Zi(G. e]g Since G0 and G1 are the bases for the 0-chains and 1-chains we have that C0(G. Since we have not yet de ned @2 we shall for the moment declare B1(G. Z2) := ker @0 = fv 2 C0(G. 1 the i-th homology with Z2 coe cients is de ned to be the quotient space Hi(G. Z2) Z5 2 C1(G. Z2 ) = 0. Z2). Z2 ): Observe that since this is a quotient space of vector spaces. d. We can nally de ne homology in this rather special setting.2. Z2) Z4 : 2 To do the computations it is convenient to use a column vector notation. 1]. Z2) j @0 v = 0g Z1(G. Example 2. Z2) = C0(G. homology with Z2 coe cients is a vector space. b. Z2 ) = Z0(G. Z2) := ker @1 = fv 2 C1(G. c = 6 1 7. Z2) C0(G.52 CHAPTER 2. Z2) := im @1 = f0 2 C0(G. For i = 0. Z2 ) such that @1 e = vg B1 (G. c. So de ne Z0(G. Z2 )=Bi(G. G0 = fa. b. d = 6 0 7. de ne the set of boundaries to be B0(G. e = 6 0 7 6 7 6 7 6 7 6 7 6 7 607 607 607 617 607 4 5 4 5 4 5 4 5 4 5 0 0 0 0 1 . Z2 ) := im @1 = fv 2 C0(G. Z2 )g Observe that B0 (G. b = 6 0 7. So let 2 3 2 3 2 3 2 3 2 3 17 07 07 07 607 617 607 607 607 6 6 6 6 607 6 7 6 7 6 7 6 7 6 7 a = 6 0 7.15 Let G be the graph representing 0. We also observed that cycles which are boundaries are not interesting. c] c.

c] = 6 7 6 7 6 0 0 1 1 74 0 5 6 0 7 5 0 4 5 4 0 0 0 0 1 Now consider Z1(G. TOPOLOGY and 53 0 0 0 1 With this convention. So H1(G. Z2 ) := ker @1 . Z2) is in Z1(G. Z2 ) if and only if @1 v = 0. the only element in Z1 (G. b. Z2) = 0. : : : 4. Thus. d. d] = 6 0 7 . b] = 6 0 7 . If we write 213 203 203 203 6 7 6 7 6 7 6 7 a. For example 2 3 2 3 1 0 0 0 72 0 3 6 0 7 61 1 0 07 6 6 7 617 6 76 7 6 7 6 0 1 1 0 7 6 1 7 = 6 1 7 = b + c: @1 b. the 0-dimensional vector space. c. Z2) = 0=0 = 0: 1 2 3 4 2 3 6 7 6 7=6 7 6 5 6 6 4 1+ 2 2+ 3 3+ 4 4 1 3 2 3 7 607 7 607 7 6 7 7 = 6 0 7.2. Observe that the vector v 2 C1 (G. @1 becomes the 5 4 matrix 2 3 1 0 0 07 61 1 0 07 6 6 7 @1 = 6 0 1 1 0 7 : 6 60 0 1 17 7 4 5 0 0 0 1 Lets do a quick check. 7 6 7 7 607 5 4 5 . c] = 6 1 7 .1. By de nition B1 (G. Z2) is 0 and hence Z1(G. Z2) := Z1(G. Z2) = 0. Z2)=B1 (G. e] = 6 0 7 : 607 607 617 607 4 5 4 5 4 5 4 5 2 6 v=6 6 4 then this is equivalent to solving the equation 1 2 3 4 3 7 7 7 5 2 6 @1 6 6 4 0 which implies that i = 0 for i = 1.

b. c] = 6 1 7 . d. Z2) Z5 =Z4 Z2: 2 2 Example 2. d. e] = 6 0 7 : 607 4 5 4 5 4 5 4 5 1 0 0 0 With this convention.16 Let G be the graph representing 1 . a]g Since G0 and G1 are the bases for the 0-chains and 1-chains we have that C0(G. Z2 )=B0(G. Z2)) Z4 . c = 6 0 7. b] = 6 0 7 . Thus. c] c. dg G1 = f a. d]. @1 becomes the 4 4 matrix 21 0 0 13 6 7 @1 = 6 1 1 0 0 7 : 60 1 1 07 4 5 0 0 1 1 Now consider Z1(G. 2 Z5 . b]. Z2 ) := ker @1 . So let 203 203 213 203 6 7 6 7 6 7 6 7 607 617 607 a = 6 0 7. Z2) thermore. d = 6 0 7 607 4 5 4 5 4 5 4 5 1 0 0 0 and 203 203 203 213 6 7 6 7 6 7 6 7 607 617 607 a. since ker @1 = 0. Z2) Z4 : 2 To do the computations it is convenient to use a column vector notation. d] = 6 0 7 . So we need to solve the equation 2 3 2 + 3 203 6 1 7 6 1+ 4 7 6 0 7 @1 6 2 7 = 6 1 + 2 7 = 6 0 7 : 6 7 6 5 4 5 4 35 4 2 37 6 7 0 + 4 3 4 . @1 (C1(G. b = 6 1 7. Then G0 = fa. We know that C0(G. MOTIVATING EXAMPLES We still need to compute H0(G. b. Z2 ) := Z0(G. Z2 ). c.54 CHAPTER 2. c. Fur2 H0(G. b. Z2) Z4 2 C1(G.

12 this is suppose to help you believe that homology might be a topological invariant. Z2 ) H (G0. TOPOLOGY Observe that since we are using Z2 coe cients. Prove that 2. Z2) Z2 . By de nition B1(G. Z2 ) where G is a graph for the following gures: 2. Z2)) Z3 . Z2) Z2 : We still need to compute H0 (G. Z2 ) := Z1(G. H1(T . Z2 ): the graph obtained by removing e and v+ from G. Z2 )=B1(G. We know that C0(G. since ker @1 Z2 . Z2 ) Z2 . Z2 ) Z4 =Z3 Z2 : 2 2 Exercises 2. Z2) = 0: In light of Proposition 2. So H1(G. Let G0 be H (G.8 Let G be a graph with a free vertex v+ that lies on edge e. Z2 ) H (G1. Z1 (G. Thus. . Z2 )=B0(G. Thus. @1(C1 (G. 2 Z4 . 1 = 2 = 3 = 4 = 1 is a non-trivial solution.2. then H (G1 G2. Of course this is not a proof of that. Z2 ) := Z0(G. Z2): 2. In particular. Z2 ) thermore. 1 55 = = = 2 3 4 is a solution. Fur2 H0(G. Z2 ) H (G2. Z2).9 Prove that if T is a tree.1. then H0(T .7 Prove that if G1 and G2 are disjoint graphs.6 Compute H (G. Z2 ) = 0.

. so we have the freedom to choose. 2. Now assume that we have two topological spaces X and Y and a continous map f : X ! Y . we can actually draw pictures of the functions.1 Approximating Maps on an Interval To keep the technicalities to an absolute minimum. either because the map is too complicated or because the dimension is too high. 2] R. The process which we adopted can be summarized as follows. With this in mind let X = p. Because we want to use the computer we can only do a nite number of evaluations of the function f . we begin our discussion with maps of the form f : a. Second. each interval can be represented by a graph and so using the types of arguments employed in the previous section we can compute the homology. we have two topological spaces and a continous map between them. 2. As was indicated in Example 2. Y = 2. We do this for two reasons. in practice we will want to apply these ideas to problems where it is not feasible to visualize the maps. Thus. We want to make sure that after we have completed all our calculations we have the correct answer. 4] R and let f : 2 X ! Y be de ned by f (x) = (x 2)(x +1). 1. This latter point is to help us develop our intuition.2. We began with a topological space G R3 which for the sake of simplicity we took to be a graph.13 there is no unique representation of these intervals as graphs. Let us begin with the representations given in Table 2.6 The question we now face is how do we go from the continuous map f . MOTIVATING EXAMPLES The purpose of the last section was to motivate the homology of topological spaces. b] ! c. In this section we will mimic this process in such a way that we obtain a linear map f : H (X ) ! H (Y ). First. to a map which takes the combinatorial data E (X ) and V (X ) to E (Y ) and V (Y )? Three issues need to be considered in constructing the map. To treat these combinatorially we think of the spaces as abstract graphs.56 2. We then observed that graphs could be represented combinatorially and nally we used this combinatorics to produce an algebraic quantity H (G) which we call the homology of G.2 Approximation of Maps CHAPTER 2. d].

3]. 3]. In the end we are only interested in computing an object f : H (X ) ! H (Y ). We include the gure to emphasize the fact that two homotopic functions can behave very di erently locally. Table 2. 4]. then f and g are homotopic. Since we want to develop algorithms that will allows us to do these computations. 0. 2]: Therefore. as will be made clear later for reasonable spaces if for every x 2 X . Let us begin with this last point. that any two functions from one interval to another are homotopic. 1].2. 1]. 1]. There are.6 we show two functions f and g which are homotopic. 2]g Vertices of X V (X ) = f 2]. 3.2. 0]. However. e. 4]g 1].2. If we move to more complicated spaces. Edges of Y E (X ) = f 2. 2]g 1]. we will do our computations in terms of edges. many di erent approaches that we could pursue. 1]. In Figure 2. 1. 0] 0. the derivatives of these functions are very di erent. so we should not need to have a very precise understanding of f but rather an approximation up to homotopy. 1] 1. 3. we want to have a systematic method for choosing which calculations to perform. 2]. 1. Recall from Exercise 2. 1]. the distance between f (x) and g(x) is su ciently small. 2. 2] and Y = 2. 0]. 1] 1. however we will adopt the following. APPROXIMATION OF MAPS Edges of X 57 E (X ) = f 2. 0]. From the combi- . 1]. 1. 0]. 2].g. then it is not true that all functions are homotopic (this is a non-trivial result).6: Edges and Vertices for the graphs of X = 2. 4]g Vertices of Y V (X ) = f 2]. of course. The second point was that we only wanted to do a nite number of calculations. 1. 0. Observe that X = 2. We have stated that homology is a homotopical invariant.

4]g. To deal with this problem let us go back to calculus to develop a method for getting good estimates on the function.5 2 Figure 2.6: The function f (x) = (x g. 1. 4] = 0. MOTIVATING EXAMPLES 3 2 1 0 -1 -2 -2 -1. and f is monotone over the edge 2. Then.17 Taylor's Theorem] Let f be a function that is n-times differentiable. 1] 1.58 4 CHAPTER 2. 2]. 3]. Since f ( 2) = 3:41421356 : : :.5 1 1. this suggests trying to map edges to edges. 1]) 0. 2. Zx n 1 (i) n 1 X f (x) = f (a) + f i!(a) (x a)i + (xn t)1)! f (n)(t) dt a ( i=1 . f ( 1) = 0.5 -1 -0. 2] 2. it is clear that f ( 2. 4]: Thus we could think of de ning a map that takes the edge 0. Theorem 2. 3] 3. 1]. this strategy of looking at the endpoints does not work for the edge 0. Of course. p 2)(x + 1) and a homotopic function natorial point of view. 1] to the collection of edges f 0. 3. 1] since f is not monotone here.5 0 0. 1].

2) reduces to f (a) 0:5jf 0(a)j 0:25 < f (x) < f (a) + 0:5jf 0(a)j + 0:25: We can think of Table 2. 1] 1. 4] = X Y: Doing this for all the edges in the domain gives the the region shown in Figure 2. 0] = 2. 2] 2.7 we have applied the relationship (2. Observe that the graph of f : X ! Y is a subset of this region and therefore we can think of the region as representing an outerbound on the function f . 0] and we can represent this graphically by means of the rectangle 0. APPROXIMATION OF MAPS 59 To apply this to our problem observe that f 00 (x) = 2 and so we can obtain the inequality Zx f (x) = f (a) + f 0(a)(x a) + (x t)2 dt a 0 (a)(x a) + (x a)2 = f (a) + f f (x) f (a) = f 0(a)(x a) + (x a)2 jf (x) f (a)j jf 0(a)jjx aj + (x a)2: For our purposes it is more convenient to write this last inequality as f (a) jf 0(a)jjx aj (x a)2 < f (x) < f (a)+ jf 0(a)jjx aj +(x a)2 : (2.2) to the midpoints of all the intervals in X and from that derived the mappings of the edges.2) implies that 1 f ( 1 ) jf 0( 1 )jjx 1 j (x 1 )2 < f (x) < f ( 1 ) + jf 0( 1 )jjx 2 j + (x 1 )2 2 2 2 2 2 2 2 1 0:25 < f (x) < 1:3713 + 0:5858 1 + 0:25 1:3713 0:5858 2 2 1:914 < f (x) < 0:8284 We can use this inequality to determine where to map the edge 0. 1]) 2. 0]: In Table 2. For example 0. 1]. Then.7 as de ning a map from edges to sets of edges. . 1]: f ( 0. let a = 2 . 1] 2. for any x 2 0.7. 0] 2. Observe that since each interval has length 1 (2.2.2.2) 1 Returning to the interval 0. 1] 7! 2. 1] the inequality (2. 1] 1. We would like to make clearer this idea of mapping edges to sets of edges.

for every x 2 X .19 Let e be and edge with endpoints v . 0]. A multivalued map F : X !Y is a ! function from X to subsets of Y . 4] if x 2 ( 2. is that we used the edges to de ne the images of the vertices. 1. 0]. then F (v) = F (e1) \ F (e2): (2. MOTIVATING EXAMPLES Image Edges Edge of X Bounds on the image 2. 1]. 0] if x 2 (0. Using this language we can view our edge mapping as a multivalued map F : 2. 1]. 2] and Y = 2.60 CHAPTER 2. Since we will used this idea later let us introduce some notation and a de nition. 0]g 1:33 < f (x) < 1:76 f 2. 1] if x = 1 > > > 2. 0) < F (x) := > 2. 1. i. 0. The corresponding open edge is e:= e n fv g: The second observation. the edges without its endpoints. 0] if x = 0 > 2. 2] 0:5 < f (x) < 3:5 f 1. 2) > > : 2. In particular. 0. De nition 2. 4] if x = 2 > > 1. 1]g 1:92 < f (x) < 0:83 f 2. 2]g Table 2. 0. 1]. 4]g 1:92 < f (x) < 0:1 f 2. 2]. 1]. 1. 1. 2] if x = 2 There are three observations to be made at this point. 2. 3. 1] 1. F (x) Y . 2] if x 2 (1.18 Let X and Y be sets. 3]. 0]. 1] if x 2 ( 1. i. 1) > > 1. 2]! 2.3) . 0] if x = 1 > > 2. we used the formula that if v is a vertex that lies in edge e1 and e2. observe that F is de ned in terms of the vertices and the interior of the edges. 1]. De nition 2. 1] 1. 1. 0] 0. 4]. 4] de ned by ! 8 1. First.e.7: Edges and Vertices for the graphs of X = 2.e. 1) > > 2.

5 0 0.5 1 1. it is completely characterized by its values on the four edges that make up X .8. 1:5 + 2 ] and Y = 12 i=0 i i 2 + 2 . p Observe that the graph of the function f (x) = (x 2)(x + 1) lies inside the graph of this edge map.2. The multivalued map F that we constructed above is fairly coarse.8. For example let us write X= 8 i=0 i i 2 + 2 .5 -1 -0. Thus. then one approach is to use ner graphs to describe X and Y . APPROXIMATION OF MAPS 4 61 3 2 1 0 -1 -2 -2 -1. The graph of the corresponding multivalued map is shown in Figure 2. 1:5 + 2 ] Using the same approximation (2.7: The graph of the map produced by sending edges to sets of edges. If we want a better approximation.2) as above we obtain the data described in Table 2.5 2 Figure 2. Observe that this is a better approximation to the function than . The nal point is that even though F : X !Y is a map that is de ned on ! uncountably many points. This is important because it means that it can be stored and manipulated by the computer.2. F is a nitely representable map.

0:5]. 3:5]g f 0:5. So in this section we begin with the question: How can we use the information in Figure 2. 2:5]. 0:5. 1. Recall that homology is by de nition a quotient of cycles by boundaries. it seems that the rst place to begin is on the level of chains. 0:5. 0:5] 0:5. 1:5]. 0]. 0:5]g f 1. 2]. 2]. 1:5]. 3:5]g f 0:5. 1]. 1:5]g f 2. 2] 1:30 < f (x) < 3:42 0:12 < f (x) < 1:46 1:53 < f (x) < 0:01 1:52 < f (x) < 0:95 1:52 < f (x) < 1:37 1:49 < f (x) < 0:83 0:95 < f (x) < 0:22 0:08 < f (x) < 1:76 f 1. we started on the level of edges. 2:5]. 1:5]. 1]. 1]. 3. 2]) is the vector space over Z2 whose basis is given by the set of . In trying to generate the algebra we will start with the vertices. 1]. Furthermore. 0]. 0:5]. 1. 2. 0] 0. 1]. 2]) ! H ( 2. In keeping with Figure 2. Recall that C0( 2. 1:5]. what was obtained with intervals of unit length. 1.2 Constructing Chain Maps In the previous section we considered the problem of approximating maps from one interval to another. 1:5. 1]. 1:5. 0:5] 0:5. 0. 1:5] 1:5. 1:5. 2:5. 2]g Table 2.7 to construct a map f : H ( 2. 3]. 2. 1:5]g f 2. 0. MOTIVATING EXAMPLES Image Edges Edge of X Bounds on the image 2. 0]. 0]. 1:5]g f 2. 1:5. one can obtain as good an approximation as one likes by choosing the edge lengths su ciently small.62 CHAPTER 2. 1. in order to be able to use our intuition from linear algebra we will consider homology with Z2 coe cients. 4]. 2] and 2. 1:5. Of course the goal of this course is to use such an approximation to reduce the analytic problem to an algebraic problem. Thus. In Figure 2. 3.8: Edges and Vertices for the graphs of X = 2. 2:5. In fact.7 we will consider 2. 1] 1. 1:5].2. In de ning the approximation. 0. 0:5]. 0:5. 2. 1] 1. 2] and Y = 2. 1. 3]. 1:5]. 1. 4])? Let us begin by emphasizing that this is not an obvious task. 1:5. 1:5]g f 0. which in turn belong to subspace of the set of chains. 0.9 one sees the graph of the multivalued map when the lengths of the edges is 0:1. 4] to be the graphs made up of the edges with vertices having integer values. 1:5] 1:5.

2]) and C0( 2. APPROXIMATION OF MAPS 4 63 3 2 1 0 -1 -2 -2 -1. 4]) is generated by the vertices ff 2g. 2]) ! C0 ( 2.5 -1 -0. f3g.5 2 Figure 2. f2g. 4]): Of course. vertices ff 2g. f0g. If we order the basis elements of C0 ( 2. f1g. f 1g. For lack of a better idea lets de ne f#0(v) := max F (v).2. We will begin by de ning a linear map f#0 : C0 ( 2. f2gg and that C0( 2. f 1g.8: The graph of the multivalued approximation to f (x) = (x p 2)(x + 1) with edges of length 0:5.5 1 1. f1g. f0g. f4gg. to de ne a linear map it is su cient to de ne how it acts on the basis elements.5 0 0.2. 4]) according to the obvious ordering of the vertices then 2 3 0 0 0 0 07 60 0 0 0 07 6 6 60 0 1 1 07 7 6 60 1 0 0 07 7 f#0 = 6 60 0 0 0 17 7 6 7 6 60 0 0 0 07 7 4 5 1 0 0 0 0 .

1]) does not map to any intervals. 3] + 3.5 -1 -0. Of course the basis of these spaces are given by the intervals.5 1 1. So consider the interval 2.64 4 CHAPTER 2. But what about f#1 ( 0. 1]) where f#1(f0g) = f#1(f1g) = f0g? Since the two endpoints are the same.5 2 Figure 2. 2]. that f#1 ( 0. 1]) = 0. We have now de ned a linear map between the 0-chains of the two spaces. The next step is to \lift" the de nition of f#0 to obtain a linear map f#1 : C1( 2. 1] 2. 4 in the obvious way an apply these . f#1 ( 1. 0]) = 0.e. MOTIVATING EXAMPLES 3 2 1 0 -1 -2 -2 -1.5 0 0. 4]. let us just declare that f#1 ( 0. 1]) = 1. 2]) ! C1( 2. Similarly. 2] and 2. Again ordering the intervals of 2. 1]. 4]).9: The graph of the multivalued approximation to f (x) = (x p 2)(x + 1) with edges of length 0:1. 1])? We know that f#0(f 2g) = f4g and f#0(f 1g) = f1g so it seems reasonable to de ne f#1 ( 2. 2] + 2. How should we de ne f#1( 2. i.

Notice that since f# is a linear map this leads to the following interesting equation @f# (c) = 0 = f#(@c): Again. Z2). in homology c] = 0. i. Thus. so the easiest constraint is to ask that f#(c) = @f# (b). How can this be guarenteed? In other words. so f (0) = 0 and hence f ( c]) = 0. we want the homology map f to be linear. But as we just noted we want f ( c]) = 0 and so the simplest condition to require is that f#(c) be a boundary which means that in homology f#(c) is in the same equivalence class as 0. Z2 ) ! H ( 2. What chain? The only one we have at our disposal is b. c = @b for some chain b. then f#(c) is a cycle. 4]. After all elements of homology are equivalence classes of cycles which are very special chains. Notice that once again we are led to the interesting equation @f# (b) = f#(c) = f#(@b): . Thus f#(c) = f#(@b). c is a boundary so we can write c = @b for some chain b. what kind of constraint on f# will guarentee that cycles which are boundaries go to boundaries? To answer this lets repeat what we have said.e. But. let c be a cycle. Now if f# is supposed to generate a map on homology. What does this mean on the level of cycles. This means that in homology c is in the equivalence class of 0. it is important that f# map cycles to cycles.e. APPROXIMATION OF MAPS 65 rules to each of the intervals we obtain the following matrix 20 0 0 03 60 0 0 07 6 60 1 0 17 7 6 7 6 f#1 = 6 1 0 0 1 7 7 6 61 0 0 07 7 4 5 1 0 0 0 In guring out how to de ne f#1 we used the phrase \it seems reasonable to de ne" but this does not mean we should not de ne it a di erent way. but this time assume that it is also a boundary. Given our choice for f#0 are there any restrictions on the way we de ne f#1? The answer is an emphatic yes. Let c be a cycle. Thus f#(c) should be a cycle which again by de nition means that @f#(c) = 0. i. by de nition @c = 0.2.2. 2]. f : H ( 2. Recall that our goal is to use f# to obtain a map on homology.e. i. our real interest is in cycles rather than arbitrary chains. If f# takes cycles to cycles. But we want f#(c) to be the boundary of some chain.

Let us now check whether the linear maps f#0 and f#1 are chain maps.2] : C2( 2. Z2) = Z2 and H1( 2. 2].2] = @1 2. MOTIVATING EXAMPLES As one might have guessed from the time spent discussing it this relationship is extremely important and linear maps on the set of chains that satisfy @f# = f#@ are called chain maps. Z2 ) = 0 . Z2) @1 2. Z2 ) ! 0: Using this notation we see that the relation @f# = f#@ should be written as f#0@1 2. 2]. Z2) ! H ( 2. that they satisfy the relation @f# = f#@ . 2].3 we know that H0 ( 2. Z2 ) ! C1 ( 2.e.2] : C1( 2. Z2 ) ! 0 and @2 2. i. 4]. Thus the maps f#0 and f#1 are chain maps. Z2) @0 2.4] : C1( 2. Z2 ) ! C0 ( 2. Z2). 2].66 CHAPTER 2. 2]. 2]. 2]. First we have two sets of boundary maps @2 2. Recall that the constraint of being a chain map was imposed in order to guarentee that f# would generate a map on homology.4]f#1: (2. From Section 2.4] : C2( 2.1. f : H ( 2. 2].4) In the matrix form this equation becomes 3 2 3 2 1 0 0 0 0 0 72 0 0 0 0 3 0 0 0 0 0 72 60 0 0 0 07 1 0 0 03 61 1 0 0 0 07 6 6 7 6 6 76 76 6 0 0 1 1 0 76 1 1 0 0 7 6 0 1 1 0 0 0 76 0 0 0 0 7 7 7 6 6 76 76 6 0 1 0 0 0 76 0 1 1 0 7 = 6 0 0 1 1 0 0 76 0 1 0 1 7 7 7 6 6 76 76 6 0 0 0 0 1 76 0 0 1 1 7 6 0 0 0 1 1 0 76 1 0 0 1 7 7 7 6 76 74 6 5 6 74 1 0 0 0 7 7 6 6 5 60 0 0 0 1 17 60 0 0 0 07 0 0 0 1 5 1 0 0 0 5 4 4 0 0 0 0 0 1 1 0 0 0 0 and it is left to the reader to check that this is an equality. 2]. Z2) @1 2.2] : C0( 2. Z2 ) ! C0 ( 2. 2]. 2]. 2]. 2].4] : C0( 2. Z2) @0 2. We were sloppy about the subscripts in our discussion above so now we need to be a bit more careful. Z2 ) ! C1 ( 2.

However. i. f0 : H0( 2. For the moment we are just trying to get a feel for how we can relate algebraic quantities to topological objects. Thus f0 (1) = 1: In other words. Since the eld Z2 consists of two elements 0 and 1. 2]. Z2) consists of two vectors which we will write as 0 and 1. We will ll in the details later. 4]. H0( 2. we can take the equivalence class which contains the vertex f 2g as a generator for H0 ( 2. we can write f 2g] = 1 2 H0( 2. Z2). Thus. 2]. Z2 ) = Z0( 2. Z2 ): How should we de ne the map f0 ? By de nition the elements of H0( 2. By looking at the matrix which represents @1 2. there is no 1-chain w such that @1 2. 2]. 2]. But @0 2.e. 4]. also show that f4g] = 1 2 H0 ( 2.e.2] it is possible to check that the vertex f 2g is not in the image of @1 2.2. Since the equivalence class of the cycle f 2g generates H0 ( 2. Z2 ) is the linear map given by multiplication by 1. Z2) ! H0( 2. they are not.e. 4]. 2]. Z2 ).2]w = f 2g. Z2 ). 2]. Exercises . 4]. Z2). Z2) = 0 Thus.2. Of course we want to use the chain map f#0 to do this. Z2 ): Returning to our map on homology. 2]. 2].2]. Z2 ) = Z2 and H1( 2. APPROXIMATION OF MAPS and similarly 67 H0( 2. Z2). i. 4]. as was mentioned above f 2g is a generator for 0 and f#0(f 2g) is a cycle so we can de ne f0 (1) to be the equivalence class which contains the cycle f#0(f 2g). 2]. the only interesting map is f0 : H0 ( 2. C0( 2.2] = 0 so any 0-chain is a 0-cycle. 1 is a homology class so f#0(1) is not de ned. to de ne f0 we need to determine f0 (1). f0 (1) := f#0(f 2g)] = f4g]: The same arguments that led to f 2g] = 1 2 H0( 2. 2]. Z2 ) are equivalence classes of the cycles Z0( 2. Z2 ) ! H0( 2. 2]. i. If you do not nd these de nitions and constructions completely rigorous that is good. Z2 ). This is probably a good place to restate the caveat that we are motivating the ideas behind homology at this point.

Discuss how to make sense of this relation as it pertains to the boundary operators on the levels of 0-chains and 2-chains. Recall that 2. 1] but where the endpoints are identi ed. i. Up to now we have considered maps from one interval to another. 2].11 we show a variety of di erent maps and indicate how many times they wrap 1 around itself.e. Our goal in this section is to see if we can detect the di erences in these maps algebraically. 0:75]. MOTIVATING EXAMPLES chains. 1] ! R given by f (x) = 2x. Z2) is well de ned. 0:25. Since the homology of an interval is fairly simple it is not surprising that the maps on homology are equally trivial. We want to think of f as a map from 1 ! 1 and do this via the identi cation of y = y + 1 (see Figure 2.68 CHAPTER 2.4) involves the boundary operators on the level of 12. 0:5].2. How should we interpret the drawing in Figure 2.3. Z2 ) ! H0( 2. going all the way around 1 in the domain results in going twice around 1 in the image. Z2 ). 0:25]. 2. it is rather di cult to draw the graph of a function f : 1 ! 1.10).4]. 0 = 1. Let us begin by considering the map f : 1 ! 1 given by f (x) = 2x(1 x) which is drawn in Figure 2. To see how this works in practice consider the function f : 0. e. In fact we will go a step further and think of 1 as the real line where we make the identi cation x = x + 1 for every x 2 R. 0:5] the graph of f covers all of 0. Of . Z2 ) = Z2 : We will focus our attention on f1 : H1( 1. @1 2. 0:5. So we divide it into the intervals 0.2] and @1 2. So going half way around 1 in the domain corresponds to going once around 1 in the image. In Figure 2. and 0:75.1. Unfortunately. 1]. 4].3 Maps of the Circle H0( 1.11(a). Z2 ) = Z2 and H1 ( 1.10 Equation (2. i. In other words. Thus. The rst step is to view 1 as a graph. 1].e. In order to draw simple pictures we will think of 1 as the unit interval 0. Z2 ) ! H1( 1. 0:5 = 1:5 = 2:5.11 Show that f0 : H0( 2. f wraps 1 around itself twice.10(b)? Observe that as we move across the interval 0. While this process allows us to draw nice gures it must be kept in mind that what we are really interested in is the f as a continuous mapping from 1 to 1 .g. So let us consider a space with non-trivial homology such as 1 of Section 2.

This implies that F (f0:25g) = F ( 0.2 0.2 0.2.8 0.4 0.2 1 1 0.1.8 1. 1] ! 0. it is easier to understand what is happening if we can view these bounds in the unit square. Thus f (0) = 0 = 1 = f (1).4 0. We do this using the Taylor approximation.6 1. Using the identi cation y = y + 1 we obtain Figure 2. Recall that we de ned the images of vertices via equation (2.4 0.3.8 0.2 0 0 0.2) becomes f (a) jf 0(a)jjx aj 2(x a)2 < f (x) < f (a) + jf 0(a)jjx aj + 2(x a)2: In Figure 2. 1] ! R.4 1.6 0.3). course 0 = 1 so this decomposition of 1 into an abstract graph is exactly the same as that used in Section 2.8 1 0 0 0.2 1.12(a) we indicate the resulting multivalued map F that is an outer approximation for f . 0:25]) \ F ( 0:25. In the right hand drawing we have made the identi cation of y = y + 1 and so can view f : 0. Since f 00(x) = 4 equation (2.8 1 Figure 2. APPROXIMATION OF MAPS (a) 2 2 (b) 69 1. It is important to keep in mind that on both the x and y axis we make the identi cation of 0 = 1. The left hand drawing indicates f : 0.6 0.6 0.4 0.6 1. 0:5]) .2.2 0. 1].10: Two versions of the graph of f (x) = 2x. The next step is to obtain an approximation for f .4 1. Of course.12(b).8 1.6 0.

8 0.8 1 Figure 2.8 1 0 0. 0:5] half way around 1 and then over the interval 0:5.2 0 0 0. (c) f wraps 1 three times around 1.8 1 (c) 1 0.2 0. Thus.4 0. One we could rede ne our multivalued map F or two we can try to make a ner approximation of 1 . Since we do not know of a more e cient way of de ning F we will adopt the approach of re ning our approximation of 1. So let us consider 1 = 0.2 0.2 0. 1] f unwraps it.8 0.6 0. = 0:25.2 0 1 0. It doesn't seems right that a connected set needs to be approximated by a disconnected set.4 0. 0:2] 0:2. 0:8] 0:8.70 (a) 1 0.11: Four di erent maps f : 1 ! 1. We have two possibilities at this point.6 0.2 0.6 0.4 0.2 0 CHAPTER 2. 0:4] 0:4.4 0. What we are saying is that using this procedure the outer approximation of a point is the union of two disjoint sets. 0:5] f0:75g: This is troubling. (d) f wraps 1 once around 1.6 0. How do these di erent f 's wraps 1 around 1 ? (a) f wraps the interval 0. but in the opposite direction from the example in (b).2 0 (d) 0 0.8 1 0 0.4 0. 0:6] 0:6. (b) f wraps 1 once around 1.6 0.6 0.4 0. MOTIVATING EXAMPLES (b) 1 0.8 0.4 0.6 0.4 0. we could say that the total amount of wrapping is 0.8 0.6 0. 1:0]: (2. This means representing 1 in terms of shorter edges.5) .

0:4] 0:8.2 -0.8 0.9 0. 0:6] 0:2.2 0. 1] if x 2 (0:8. 0:6] 0:4. 0:4] 0:2. 0:6] 0:4. 1] if x = 1 . Using this approximation F (v) is an interval for every vertex v. Using the same rules as before we end up with the multivalued map 8 > > > > > > > > > > < F (x) = > > > > > > > > > > : 0. 0:4] 0.2 0. 1) 0:8.1 -0. 0:4] 0. 0:4] 0.8 1 0 0 0.13.6 0.3 0. 1] if x 2 (0.6 0.2.4 0 0.2 0.8 1 Figure 2. 0:2) if x = 0:2 if x 2 (0:2.6 0. 0:6) if x = 0:6 if x 2 (0:6.8 0. 0:6] 0:4.5 0. APPROXIMATION OF MAPS (a) 1 1 (b) 71 0.4 0 0. 0:6] 0:2. 1] if x = 0 0:8.12: The outer approximation for the map f (x) = 2x(1 x).4 0. 0:8) if x = 0:8 0:8. 0:4) if x = 0:4 if x 2 (0:4.4 0.4 0. If we repeat the approximation scheme described above for this representation of 1 we get the outer approximation described in Figure 2. 0:4] 0:2.2 0.2.7 0.6 0.

5). 0:4] + 0:4.2 0.1.1.2 0. The reader is encouraged to check that in this case the homology of 1 does not change. Of course.8 1 0 0 0. However. 1:0]] = 1 2 H1 ( 1.6 0. 0:8] + 0:8. Z2) ! C1( 1 . Z2) in the same manner as in Section 5. MOTIVATING EXAMPLES (b) 1 0.2 -0.4 0.5 0.1 0.6 0.3 0 0. Set f#0(v) = max F (v) for any vertex v. what should be clear is that it would be nice to have a general theorem that says that if one has the homology of a space does not depend on the approximation used in the computation.2 0 0.72 (a) 0. f#0(f0g) = f1g and f#0 (f0:2g) = f0:4g.1 -0. 0:2] + 0:2.1 0. For the moment we will just assert that the 1-chain given by the sum of all the intervals generates H1( 1.8 1 Figure 2. Z2) ! C0( 1 . the construction of f#1 : C1( 1 . Thus for example.e.13: The outer approximation for the map f (x) = 2x(1 x) based on edges of length 0:2. Z2 ): Having determined the multivalued map F for this approximation we will construct the chain map f#0 : C0 ( 1.6 0.4 0.7 0.2 0.8 0.4 0. i.6 CHAPTER 2. Again. Having de ned f#0. Z2) also follows as in Section 5. 1 0. Z2 ).5 0.4 0. .3 0. we have not computed the homology of the graph representing given by (2. we will address these issues later. 0:6] + 0:6.9 0.

Z2 ) ! H1( 1. Z2).14 shows the resulting multivalued map. Z2 ) is given by multiplication by 0. 0:4]+ 0:4. Notice that this corresponds to the number of times that f wraps 1 around its. . APPROXIMATION OF MAPS 73 Using the natural ordering of the intervals which are a basis for C1( 1. 0:4] + 0:4. We proceed exactly as before. 1:0]) is given by 2 61 61 6 60 6 60 4 0 0 1 0 0 0 0 0 0 0 0 0 0 1 0 0 1 1 0 0 0 32 3 2 3 2 3 76 1 7 6 2 7 6 0 7 76 1 7 6 2 7 6 0 7 76 7 6 7 6 7 76 1 7 = 6 2 7 = 6 0 7: 76 7 6 7 6 7 76 1 7 6 0 7 6 0 7 54 5 4 5 4 5 1 0 0 Therefore. In vector notation as an element of C1( 1 . Z2 ). 0:2]+ 0:2. Z2 ) we need to see how f#1 acts on the generator of H1 ( 1. To obtain an appropriate multivalued map we have chosen to represent 1 as follows 1 = 0. 0:625] 0:625. Since f 00 (x) = 2 we can use equation (2.2. f1 : H1( 1. 0:375] 0:375. Z2). Figure 2. 0:875] 0:875. 0:8] + 0:8. Again we need estimates on the approximation. 0:125] 0:125. 0:5] 0:5. 0:75] 0:75. 0:6]+ 0:6.2. 1:0] = 6 1 7 : 6 7 617 4 5 1 Recall that we are using Z2 coe cients hence f#1 ( 0. 0:6] + 0:6. Lets do this again for the map f (x) = x2 . 1] As before it is the sum of all these intervals which generates H1( 1 . we have 2 3 617 617 6 7 0.2). 0:25] 0:25. 0:8] + 0:8. 0:2] + 0:2. Z2 ) we can write 2 3 1 0 0 0 17 61 0 0 0 17 6 6 7 f#1 = 6 0 1 0 1 0 7 6 60 0 0 0 07 7 4 5 0 0 0 0 0 In order to understand the induced map on H1 ( 1.

MOTIVATING EXAMPLES (b) 1 0.4 0.2 0.74 (a) 1.2 CHAPTER 2.6 0.1 -0.2 0 0.6 0.4 0. then we are .8 0.6 0.2 0 0.8 1 Figure 2.4 0.3 0.9 1 0.8 0. Constructing f# as before and using the natural ordering of the intervals which are a basis for C1 ( 1.6 0.5 0.2 0.14: The outer approximation for the map f (x) = x2 .4 0. Z2) we can write 2 1 60 6 60 6 6 6 f#1 = 6 0 60 6 6 60 6 60 4 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 1 3 7 7 7 7 7 7 7 7 7 7 7 7 7 5 If we let f#1 act on the 1-chain which generates H1( 1. Z2 ).8 1 0 0 0.7 0.2 0.

Z2 ) is given by f1(1) = 1.e. 0:125] 0:125. 1]: Following exactly the same process as in the case of f (x) = x2 we obtain 21 0 0 1 0 0 0 13 61 0 0 0 1 0 0 07 6 7 61 0 0 0 1 0 0 07 7 6 6 7 60 1 0 0 0 1 0 07 7 f#1 = 6 0 1 0 0 0 1 0 0 7 6 7 6 6 7 60 0 1 0 0 0 1 07 6 7 60 0 1 0 0 0 1 07 5 4 0 0 0 1 0 0 0 1 Again viewing how this acts on the generator of H1 ( 1. it is multiplication by 1. i. 0:25] 0:25. 0:375] 0:375.2. f1 : H1 ( 1. We shall do one more example. Z2 ) ! H1( 1. 0:875] 0:875. 0:5] 0:5. 0:75] 0:75. Figure 2. Z2) we have 2 1 0 0 1 0 0 0 1 32 1 3 2 2 3 2 0 3 6 1 0 0 0 1 0 0 0 76 1 7 6 2 7 6 0 7 6 76 7 6 7 6 7 6 1 0 0 0 1 0 0 0 76 1 7 6 2 7 6 0 7 76 7 6 7 6 7 6 6 76 7 6 7 6 7 6 0 1 0 0 0 1 0 0 76 1 7 6 2 7 6 0 7 6 76 7 6 7 6 7 6 0 1 0 0 0 1 0 0 76 1 7 = 6 2 7 = 6 0 7 76 7 6 7 6 7 6 6 76 7 6 7 6 7 6 0 0 1 0 0 0 1 0 76 1 7 6 2 7 6 0 7 6 76 7 6 7 6 7 6 0 0 1 0 0 0 1 0 76 1 7 6 2 7 6 0 7 54 5 4 5 4 5 4 0 2 1 0 0 0 1 0 0 0 1 . APPROXIMATION OF MAPS 75 performing the following computation 2 1 0 0 0 0 0 0 0 32 1 3 2 1 3 6 0 0 1 0 0 0 0 0 76 1 7 6 1 7 6 6 0 0 0 1 0 0 0 0 76 1 7 6 1 7 76 7 6 7 6 76 7 6 7 6 6 0 0 0 0 1 0 0 0 76 1 7 6 1 7 76 7 6 7 6 6 0 0 0 0 0 1 0 0 76 1 7 = 6 1 7 76 7 6 7 6 76 7 6 7 6 6 0 0 0 0 0 0 1 0 76 1 7 6 1 7 76 7 6 7 6 6 0 0 0 0 0 0 1 0 76 1 7 6 1 7 76 7 6 7 4 54 5 4 5 0 0 0 0 0 0 0 1 1 1 Thus.15 shows the multivalued map that acts as an outer approximation when the represention of 1 is given by 1 = 0.2. 0:625] 0:625. Observe that this again is the same as the number of times that f (x) = x2 wraps 1 around itself. that of f (x) = 2x.

.e. On the other hand.5 0.4 0. MOTIVATING EXAMPLES (b) 1 0. With this in mind we will spend the next chapter studying the algebra needed to be able to rigorously do homology over the integers.6 0. In this case we end up with f1 (1) = 0.9 2 0. Unfortunately. This does not match our geomtrical observation that f (x) = 2x wraps 1 around itself twice.5 0. the homology map on the rst level is multiplication by 0. If we had been using integers we might expect to obtain that f1 is multiplication by 2.2 0. i.8 1 0 0 0.5 CHAPTER 2.3 0. it is clear that f1(1) = 0 precisely because we are using Z2 coe cients.4 0.1 -0. Z2) for f (x) = 3x.8 1 Figure 2.8 0.2 0 0.5 0.4 0.7 1. Exercises 2.5 0 0. using the integers as a scalar does not lead to a vector space.6 1 0.6 0.2 0.12 Compute f1 : H1( 1.15: The outer approximation for the map f (x) = 2x. Z2 ) ! H1( 1 .76 (a) 2.

b.b).g. e. There exist an identity element 0 2 G such that for all a 2 G a + 0 = 0 + a = a: 3.Chapter 3 Abelian Groups In Chapter 2 we computed homology groups using linear algebra. c 2 G. For each a 2 G there exists an inverse a 2 G such that a + a = b + b = 0: 77 . As was pointed out in our analysis of maps on the circle it would be nice if we could move beyond linear algebra. In this chapter we will introduce the abelian group theory that lies at the basis of homological algebra. For all a. q(a. a + (b + c) = (a + b) + c (associativity) 2. 3. one typically uses a notation such as a + b or ab. Rather than writing the operation in this functional form. together with a binary operation + de ned on G and satisfying the following four axioms: 1.1 An abelian group is a set G.1 Groups De nition 3. A binary operation on a set G is any mapping q : G G ! G.

2 We denote the set of integers by Z. Example 3. Addition is a binary operation on N . 2. associative and 0 2 N. 0 2 H . 1 2 N. it is commutative. i. 0). A nonempty G is a subgroup of G if: 1.e. All these sets are abelian of nonnegative integers. but 1 62 N. 1. Example 3. However. Example 3.1) that the identity element 0 is unique and that given any a 2 G its inverse element a is also unique. for every a 2 H its inverse a 2 H . b 2 G. It is convenient to describe nite groups such as Zn by giving their table of addition. b) ! (a + b) mod n.3 Recall that the set N of natural numbers is the same as the set Example 3.5 Given a positive integer n. For example. CHAPTER 3. : : :n 1g with the addition de ned by (a. 2.78 4. the rationals by Q. . groups under addition.4 The vector space Rn is an abelian group under coordinatewise addition with the identity element 0 = (0. 0. here is one for Z3 : + 0 1 2 0 0 1 2 1 1 2 0 2 2 0 1 subset H De nition 3.6 Let G be a group with the binary operation +. We shall abandon the mod n notation when it will be clear that we mean the addition in Zn and not in Z. ABELIAN GROUPS a+b=b+a (commutativity) It follows from the axioms (see Exercise 3. the smallest integer c 0 such that a + b c is divisible by n. N is not an abelian group since its elements have no inverses under addition. the real numbers by R and the complex numers by C. : : : . where (a + b) mod n is the remainder of a + b 2 Z in the division by n. For all a. Furthermore. let Zn := f0.

b 2 H . If n is a negative integer. X aj gj : (3.e. 79 Proposition 3. Observe that for any k 2 Z n f0g. H is closed under +.3. The proof of this proposition is left as an exercise. Given a 2 G and n 2 Z. . GROUPS 3. given a. What makes vector spaces so nice is that they have bases which one can use to uniquely represent any vector in the vector space.1) X aj gj where aj 2 Z. then this should be interpreted as the n-fold sum of a.8 Given a group G. b 2 H . a + b 2 H .2) Example 3. De nition 3. However. Then H is a subgroup of G. we use the notation na := a + a + + a | {z } n terms to denote the sum of a with itself n times. De nition 3. if k = 1. This is easily seen by noting that if fkg did generate. then fkg does not 6 generate Z.1. The elements of fgj gj2J are called generators. i. a set of elements fgj gj2J G if any a 2 G can be written as a nite sum a= G generates (3. implies a b 2 H . fkg is a maximal linearly independent set.7 Let H be a subset of G with the property that for any a. then there would be an integer n such that nk = 1. If there is a nite set of generators. then G is a nitely generated group. By the niteness of the above sum we mean that aj = 0 for all but nitely many j . Observe that the concept of a generating set for a group is similar to that of a spanning set in linear algebra.10 The group of integers Z is a free group generated by a single element basis: either f1g or f 1g.9 A family fgj gj2J of generators is called a basis of G if for any a 2 G there is a unique set of integers aj such that a= A group is free if it has a basis.

.. To see this assume that fgj gj2J formed a basis for Q. X 0 = aj gj . gng and fh1 ... 2 6 6 6 6 4 h1 h2 . Thus 2 6 6 6 6 4 h1 h2 .12 Any two bases of a nitely generated free abelian group G have the same number of elements. g2 . hm g be two bases of G with n < m... aj = 0 for all j 2 J: Every vector space has a basis. : : : . Let fg1.. j 2J Example 3. Now observe that p1 p2 = (p2 q1)p1 =q1 = (p1 q2)p2 =q2 which violates the uniqueness condition.. Recall that any element a 2 Q can be written in the form a = p=q where p and q are relatively prime integers. Proof: The proof is by contradiction. hm 3 2 7 6 7 7 = AB 6 6 7 6 5 4 h1 h2 . hm 3 7 7 7: 7 5 . Then each element of one basis can be expressed as a linear combiation of the elements of the other basis with integer coe cients. 7 5 where A = (aij ) and B = (bij ) are. 5 4 gn 3 2 7 6 7 7=B6 6 7 6 5 4 h1 h2 . Theorem 3. This number is called the rank of G. this is not true for groups. By using matrix notation. there exists p1 =q1 and p2 =q2 in the basis.. Assume that the basis consists of a unique element g = p=q. In particular. but it is impossible to solve the equation ng = g=2 for some integer n. i. Therefore. hm 3 2 7 6 7 7 = A6 6 7 6 5 4 g1 g2 . the basis must contain more than one element.80 CHAPTER 3.e.. h2 . respectively. : : : . ABELIAN GROUPS Observe that the uniqueness condition implies that a set of generators fgj gj2J of a group G is a basis of G if and only if it is linearly independent. m n and n m matrices with integer coe cients. gn 3 2g 7 6 g1 7 7 and 6 .11 The group of rational numbers Q is not free..2 6 7 6 . hm 3 7 7 7. Then g=2 2 Q.

The order of G denoted by jGj is the number of elements of G. This set is a basis for 2Z 3Z which is a proper subgroup of Z2 of the same rank 2. 0) and j := (0. and 1 if not.e. If a is of in nite order. Set i := (1. it has no elements of nite order. Example 3. 1 and 5 have order 6 thus each of them generates the whole group. j ig.14 The addition table for Z6 is as follows: + 0 1 2 3 4 5 0 0 1 2 3 4 5 1 1 2 3 4 5 0 2 2 3 4 5 0 1 3 3 4 5 0 1 2 4 4 5 0 1 2 3 5 5 0 1 2 3 4 Using the table it is easy to check that: 0 has order 1. GROUPS 81 By the uniqueness of the expansion. Thus jZj = 1 and jZnj = n. Note the relation between the divisors of 6 and orders of elements of Z6 .3. 2 has order 3 and 3 has order 2. the ranks of A and B are at most n. 3jg is not a basis for Z2 even though it is a maximal linearly independet set in Z2. Example 3. the cyclic group hai is a free abelian group which may also be denoted by Za or by aZ. Then. In general. The set of all elements in G with nite order is a subgroup called the torsion subgroup of G. fi.13 Consider the group Z2 . But f2i. We end this section with the following observation . Of course. Observe that jhaij = o(a). A group G generated by a single element a is called cyclic and is denoted by hai. i. a group which has a cyclic element of nite order other than zero cannot be free. thus the rank of AB is at most n. if a 2 G then hai is a cyclic subgroup of G. We will learn more about product groups in the next section. AB = 1m m (the identity m m matrix) which contradicts that n < m: Indeed. 1). jg is a basis for Z2 and so the rank of Z2 is 2. Observe that a free group is torsion free. But the rank of 1m m is m > n. The converse is not true (see exercises). if it exists. The order of an element a 2 G denoted by o(a) is the smallest positive integer n such that na = 0.1. Another choice of basis is fi.

there exists h 2 H of the form h = (nk + r)a where 0 < r < k. Z8 .3 (a) Determine the orders of all elements of Z5 . (b) Prove that.2 (a) Write down the tables of addition and multiplication for Z5 . (a) Show that if G is free then it is torsion-free. Z6 . Z6. 3. given any a 2 G.82 CHAPTER 3. Z8 (b) Determine the orders of all elements of Z5 . (c) Let now Zn := fk 2 Zn : k and n are relatively primeg. show that Z05 is a multiplicative group but Z06 . we get ra 2 H . (b) If Z0n := Zn n 0.4 Prove Proposition 3.1 Let G be a group. Let k be the smallest positive integer such that ka 2 H . Z6.5 Let G be an abelian group. We show that ka generates H . . ABELIAN GROUPS Lemma 3. (c) Show that if G is nitely generated and torsion free then it is free. Z8 . Prove that H is a subgroup of G. Exercises 3. Show that Zn is a multiplicative group for any positive integer n.7 3. 3. Indeed. (b) Show that the additive group Q is torsion-free. Since nka 2 H . where Zn is de ned in the preceeding exercise and the order of a in a multiplicative group is the least positive integer n such that an = 1 3. Clearly. Proof: Let G be a cyclic group generated by a and let H 6= 0 be a subroup of G. (a) Let H := fa 2 G j o(a) < 1g f0g. nka 2 H for all integers n and we need to show that all elements of H are of that form. which contradicts the minimality of k. Z08 are not.15 Any subgroup of a cyclic group is cyclic. (a) Prove that the identity element 0 is unique. if not. the inverse a of a is unique.

16 Let G be the sum of its subgroups A and B . an) + (b1 . a2 2 A and b1 . b 2 B g : We say that G is a direct sum of A and B and write (3. an + bn ) called the direct product of G1. a1 . : : : . bn) = (a1 + b1 . : : : . c = 0. a2. b2. Gn. b := c. We have the following simple criterion for a direct sum.2. : : : . Proposition 3. Then G = A B if and only if A \ B = f0g. We de ne their sum by A + B := fc 2 G : c = a + b for some a 2 A. G becomes a group with the coordinate-wise addition (a1 . There is an obvious analogy between the addition and scalar multiplication in the vector space Rn and in the direct product of groups: the di erence is that in the direct product of groups we are only allowed to multiply by integer scalars from Z. Gn of subgroups of a given group G. let A \ B = f0g and let c 2 A \ B . a2 + b2 . Then c can be 6 6 decomposed as c = a + b in at least two ways: by posing a := c. G is the direct sum of . G2. Then a1 a2 = b2 b1 2 A \ B = f0g hence a1 = a2 and b1 = b2 . In a similar way one de nes the sum and direct sum of any family G1 . b := 0 or a := 0. : : : . Hence the decomposition is unique. Conversely. Gn. G2.4) G := A B if G = A + B and the decomposition c = a + b of any c 2 G is unique. : : : . Proof: Suppose that A \ B = f0g and that c = a1 + b1 = a2 + b2 are two decompositions of c 2 G. b2 2 B . : : : . G2. PRODUCTS AND SUMS 83 G= n Y i=1 Gi = G1 G2 Gn (3.3.3) be the cartesian product of G1 . Let A and B be subgroups of G. The direct product of n copies of a group G is simply denoted by Gn. : : : .2 Products and Sums Let G1. Gn be a family of groups and let 3. G2.

2. In this text. : : : . eng. By the de nition of a basis.1 for a sum to be a direct sum is Gi \ Gj = f0g if i 6= j : There is a close relation between direct products and direct sums. ABELIAN GROUPS G1. G = Zg1 Zg2 Zgn : Example 3. The same consideration applies to Zn with the canonical basis fe1 . Example 3. however. . many direct sum decompositions. Gn if every g 2 G can be uniquelly written as a = Pn=1 gi. When in nite families of groups are considered. g2.g. G2. Then Z2 = Zi Zj. Let G be a free abelian group with a basis fg1. : : : . n. gng. where the coordinates of ei are given by ( i ) = 1 if i = j . their direct sum may only be identi ed with a subgroup of the direct product consisting of sequences which have zeros in all but nitely many places. (e j 0 otherwise. and hence. we may write G = G1 G2 Gn. jg called the canonical basis of Z2.5) 84 i=1 The criterion analogous to that in Proposition 2. hence we may write Z2 = Z Z. e2. there may be many bases. : : : . We write n M G = Gi = G1 G2 Gn : (3. Let G = G1 G2 Gn. where i gi 2 G for all i = 1.CHAPTER 3. e. we shall not need to study in nite sums and products. We may identify each Gi with the subgroup ji Gi := f0g f0g Gi f0g : |{z} f0g i'th place Then G = j1 G1 j2G2 jn Gn and. This identi cation of direct products and sums will become more formal when we talk about isomorphisms of groups in the next section.17 As for vector spaces. Z2 = Zi Z(j i). for the simplicity of notation.18 Consider the group Z2 = Z Z. This decomposition of Z2 to a direct sum is related to a particular choice of basis fi. : : : .

o((1. With the discussion of Chapter 2 in mind. 2. Note that if S = f1. o((1. 1) have order 2.19 In the group Z2 2 = Z2 Z2 of order 4. 0)) = 2.18. 2. an are integers. :::ng we recover ZS = Zn with the canonical basis ei de ned in Example 3. ^ It is easily veri ed that S := fs1. y. a set of edges or vertices in a graph. '(2) = y. s^ng is a basis for ZS . let us go back to the de nition of cartesian product in (3. Thus this is not a cyclic group. 2). or as is more relevant to this course. The notion of isomorphism introduced in the next section will permit to identify this group with Z6 . '(3) = z. : : : . : : : . 3g ! G given by '(1) = x. the goal is to give meaning to the sum Example 3. : : : . It is called the ^ ^ canonical basis and it may be identi ed with S . sng be any nite set of objects. n ^ de ned by ( si(sj ) := 1 if i = j . 1). (1. 1)) = o((0. s2. For example. 2.3. Let S = a1s1 + a2 s2 + + ansn . The group structure is given by pointwise addition: (' + )(i) := '(i) + (i). 0). With the understanding of this we may now de ne the free abelian group ZS generated by S as the set of all functions ' : S ! Z. z) 2 G3 formally is a function ' : f1.17 will now be approached in a di erent way. : : : . 2)) = 3. ng to G. a2 . :::n : Why is this a free group? Consider the functions si : S ! Z. i = 1. Thus a point (x. 2)) = 6 : Thus Z2 Z3 is cyclic of order 6.3). ^ 0 otherwise. o((0. The cartesian product Gn of n copies of G formally is the set of all functions ' from the nite set f1. Exercises . The same consideration applies to Zn Zm where n and m are relatively prime (see exercises). For this purpose. 2. What the objects are does not matter. s2. generated by (1. and (1. with the pointwise addition (' + )(si) := '(si) + (si ). i = 1. Here are the orders of its elements: 85 Example 3. 2. all 3 nonzero elements o(0) = 1. S may be a class of mathematics students. PRODUCTS AND SUMS (0. 1)) = o((1. 1) and by (1.2. where a1. : : : . Consider the group Z2 Z3 . fs1.

.6 (a) Let m. (a) If G is a nite multiplicative group and a 2 G. n be relatively prime.4 with H = hai) (b) (Fermat's Little Theorem) If p is a prime and and p does not divide a 2 Z then ap 1 1 (mod p) : (Hint: Recall Exercice 2(c) Section 1) (c) If p is a prime then bp b (mod p) for all b 2 Z. (Hint: Use Proposition 2. then ajGj = 1.86 CHAPTER 3. (b) Prove that if G is a cyclic group and p is a prime dividing jGj. Show that Zm Zn is cyclic of order mn. 3. 3.8 Prove the following statements. (b) Let G = Z12 Z36 . Express G as a direct sum of cyclic groups whose orders are powers of primes.7 (a) Prove that a group of prime order has no proper subgroup. then G contains an element of order p. ABELIAN GROUPS 3.

It is also easy to check that this operation is associative and commutative. b + h = a + h1 h2 + h 2 a + H so b + H a + H . in other words does it depend on which representative we use. in the setting of vector spaces we de ned homology as a quotient of chains by boundaries. a 2 G and b = a + h0. if b a 2 H then b + 0 = a + (b a) 2 (b + H ) \ (a + H ). Hence.3..6) Observe that 0] + a] = 0 + a] = a] so 0] acts like an identity element. so there are inverse elements. The reverse inclusion holds by the symmetric argument. so we shorten it to a] := a + H . We can de ne a binary operation on the set of cosets by setting a] + b] = a + b]: (3. Then b a = h2 h1 2 H . (b) Let a+H = b+H and let h1 . Then there exist h1 . for any h 2 H . The element a is called its representative.3 Quotients In Chapter 2. Writing cosets in the form of a + H is a bit cumbersome. h2 such that a + h1 = b + h2 . (b) a + H = b + H if and only if b a 2 H . b 2 G. Let H be a subgroup of G and a 2 G.3.20 Let H be a subgroup of G and a. For example. Furthermore. Proposition 3. The following proposition makes this precise. Notice that to use this notation it is essential that we know the subgroup H that is being used to form the cosets. Conversely. thus the concusion follows from (a). a] + a] = a + a] = 0]. then a and b are representatives for the same coset. . The only serious issue is whether this new operation is well de ned. The set a + H := fa + h : h 2 H g is called a coset of H in G. We need to extend this idea to the setting of groups. QUOTIENTS 87 3. Proof: (a) Suppose that (a + H ) \ (b + H ) 6= . Typically a coset will have many di erent representatives. Then (a) The cosets a + H and b + H are either equal or disjoint. let h0 2 H . h2 be as in (a).

for all a 2 G. Example 3. Proof: If a0 + H = a + H and b0 + H = b + H then.22 The group of cosets described by Proposition 2. : : : . the group G=H = Z=kZ has k elements 0]. Proposition 3. Thus. de nes a group structure on fa + H ga2G. will become clear in the next section. De nition 3. ii) a b . ABELIAN GROUPS Proposition 3. The equivalence class of a 2 G is the set of all b 2 G such that b a. b. 1]. Hence a0 + b0 + H = a + b + H . when we talk about isomorphisms. An alternative way of introducing the quotient group is in terms of an equivalence relation. h1 = h2. this group may be identi ed with Zk discussed in the previous section. and therefore. by Proposition 3. a0 a 2 H. k 1].2(a). iii) a b and b c ) a c.23 Let G be a nite group and H its subgroup. Then each coset a + H has the same number of alements. by Proposition 3. i. Consequently.3 is called the quotient group of G by H and denoted by G=H . jGj = jG=H j jH j : Proof: The rst conclusion is an obvious consequence of the cancellation law for the group addition: a + h1 = a + h2 .88 CHAPTER 3.e.24 Let G = Z and H = kZ for some k 2 Z. k 6= 0. for all a.20. b a. b0 b 2 H and so (a0 + b0 ) (a + b) = (a0 a) + (b0 b) 2 H . c 2 G. Since the coset . What \identi cation" means. a + b] is also represented by the remainder of the division of a + b by k. De ne the relation a b if and only if b a 2 H .21 The formula (3.6) does not dependent on the choice of coset representative used. The second conclusion is an immediate conseqence of the rst one and the Proposition 2. for all a. i) a a. b 2 G. Note that this is an equivalence relation in G.20 the group of cosets exactly is the group of equivalence classes of a 2 G.

1). we have G=H = f mi]gm2Z. b 2 G. There are some immediate consequences of this de nition. in other words they need to respect the binary operation. n 2 Z and the representatives may be sarched. n) 2 Z2 can be written as (m + n)i + n(j i) 2 (m + n)i + H . This leads to the following de nition. Example 3. 2 0. It is easily seen that ki] = mi] whenever k = m. 1] by gluing 1 to 0. f (0) = f (0 + 0) = f (0) + f (0) f (0) f (0) = f (0) 0 = f (0) . R=Z is represented by the points of that interval. If we wish to compare them then we need to be able to talk about functions between them. 1) by the fractional part of + . 1). De nition 3. HOMOMORPHISMS 89 any real number is an integer translation of a number in the interval 0. For any .4. Example 3. y = sin . Since 1 0. the coset + ] is represented in 0. Moreover there is a bijection between R=Z and 0. Thus the elements of R=2 Z may be identi ed with the points on the circle x2 + y2 = 1 in the plane. 6 6 thus there is a bijection between G=H and Z. R=Z may be visualised as a circle obtained from the interval 0. via the polar coordinate in x = cos .4 Homomorphisms Let G and G0 be two abelian groups.25 Let G = Z2 and H = Z(j i) = f( n. We may choose coset representatives of the form mi = (m. Since any element (m. in the interval 0. The equivalence relation is now . 1).27 A map f : G ! G0 is called a homomorphism if f (a + b) = f (a) + f (b) for all a. 0). For example. 2 ). as the following argument shows.3. m 2 Z. n) : n 2 Zg. for example. A very similar example explaining the concept of polar coordinates is the quotient group R=2 Z. since no two numbers in that interval may di er by an integer. Of course these functions need to preserve the group structure. Since 3.26 Consider Z as a subgroup of R and the quotient R=Z. = 2n . homomorphisms map the identity element to the identity element.

Finally. Let b 2 H . Proof: (a) We must show that f (H ) satis es the group axioms. the binary operation on f (H ) is the same as that of G0 and therefore is associative and commutative. Furthermore.30 A homomorphism f : G ! G0 is called an epimorphism if it is surjective (or onto) i. b + b0 = f (a) + f (a0 ) = f (a + a0) 2 f (H ). Then (a) for any subgroup H of G. f (a) = f (a).90 CHAPTER 3. its image f (H ) is a subgroup of G0 . ABELIAN GROUPS f (na) = nf (a) A similarly trivial argument shows that for all n 2 Z and a 2 G. Since f (H ) G0.e. (b) for any subgroup H 0 of G0. 0 2 H . a0 2 H such that f (a) = b and f (a0) = b0 . . by the previous proposition is a subgroup of G0. Now observe that 0 = f (a + a) = f (a) + f ( a). This condition obviously is equivalent to the condition ker f = 0. Therefore. De nition 3. b0 2 f (H ).29 The set im f := f (G) is called the image or range of f in G and. (c) if f is bijective (i. for any a 6= b in G. Finally.e. then there exists a 2 G such that b = f (a). then there exist a. (b) and (c) follow from similar types of arguments and are left to the reader. we need to show that f (H ) is closed under the operation +.28 Let f : G ! G0 be a homomorphism. De nition 3. one-to-one and onto) then its inverse f 1 : G0 ! G also is a bijective homomorphism. im f = G0 and a monomorphism if it is injective (or 1-1). f is called an isomorphism if it is both a monomorphism and an epimorphism. Proposition 3. its inverse image f 1 (H ) is a subgroup of G. Since f (0) = 0.e. i. The set ker f := f 1(0) = fa 2 G j f (a) = 0g is called the kernel of f and is a subgroup of G. f (a) 6= f (b). If b.

Thus. It is easy to show that f is invertible if and only if it is bijective. Then f : Z ! G de ned by f (n) = na is an isomorphism with the inverse de ned by f 1(na) = n. in general. unless an additional structure that is not preserved by isomorphisms is involved. . called inverse of f with the property gf = 1X and fg = 1Y (3. Example 3. This can be generalised to direct sums and products of any nite number of groups. When we speak about a particular class of maps. b) where c = a + b is the unique decomposition of c 2 G with a 2 A and b 2 B . the continuity of its inverse. by an invertible map or an isomorphism we mean a map which has an inverse in the same class of maps. For example. By the same argument. we may then write f : G ! G0 or G = G0. if there exists f ' an isomorphism f : G ! G0. Groups G and G0 are called isomorphic.7) where 1X and 1Y denote the identity maps on X and Y respectively. Proposition 3. any cyclic group of order k is isomorphic to Zk . an isomorphism would be a continuous map which has a continuous inverse: The continuity of a bijective map does not guarantee. It is easy to see that G = G0 is an equivalence relation. HOMOMORPHISMS 91 The last de nition requires some discussion since the word isomorphism takes di erent meanings in di erent branches of mathematics. a homomorphism is an isomorphism if and only if it is invertible in the class of homomorphisms. Example 3.3. Let X. Y be any sets and f : X ! Y any map.4. inverse de ned by f 1(c) = (a. If this is the case.33 Let G be a cyclic group of in nite order generated by a.31 Z6 = Z2 Z3 . When G = G0.1(c) guarantees that this problem does not occur in the class of homomorphisms.32 Let A. Then f is called invertible if there exists a map g : Y ! X . Then the map f : A B ! G de ned by f (a. We shall often permit ourselves to identify isomorphic groups. g is uniquelly determined and denoted by f 1. b) = a + b is an isomorphism with the Example 3. a homomorphism f : G ! G may be also be called an endomorphism and an isomorphism f : G ! G may be called an automorphism. if continuous maps are of concern. notation G = G0. B be subgroups of G such that G = A B .

the only automorphisms of Z are 1Z and 1Z . B = Zj of Z2 . for any a 2 G and h 2 H . we have f (a + h) = f (a). We may now state the following the map Theorem 3. if a + H = b + H then b a 2 H thus f (b) = f (a). the map f : Zn ! G de ned on the elements of the canonical basis by f (ei) := si and extended by linearity is a well de ned isomorphism. is an epimorphism.92 CHAPTER 3. completely de ned by its values on 1. f is trivial and ker f = Z. take subgroups A = Zi. and G be as above. : : : . called the quotient isomorphism.35 Let f : Z ! Z be any homomorphism. Another choice of a left inverse of i is p0 (ni + mj) = (n + m)i (a "slant" projection). Then. Note that pi = 1A hence p may be called a left inverse of i and i a right inverse of p. B . s2. Indeed. Example 3.36 Let A. Example 3. If f (1) = k then f (n) = nk for all n. It is easy to see that q is an epimorphism and its kernel is precisely H . This map is called the canonical quotient homomorphism. that image is independent on the choice of a representative of a coset a + H . Indeed. Indeed. sng discussed in the previous section. .34 Let G be a free abelian group generated by fs1. Then G = Zn.37 Let H be a subgroup of G and de ne q : G ! G=H by the formula q(a) := a+H . Otherwise ker f = 0 and im f = kZ. Since kZ = Z if and only if k = 1. The inclusion map i : A ! G is a monomorphism and the projection map p : G ! A de ned by p(c) = a where c = a + b with a 2 A and b 2 B . Let now f : G ! G0 be a homomorphism and H = ker f . By linearity. If k = 0. f is Example 3. ABELIAN GROUPS Example 3. Then f : G=H ! im f de ned by f (a + H ) = f (a) is an isomorphism. Note that a left inverse is not necessarily unique. Hence the image of any coset a + H under f is f (a + H ) = ff (a)g : Moreover.38 Let f : G ! G0 be a homomorphism and H = ker f .

i the primitive square root of 1. the formula for f is independent of the choice of coset representatives. Let S 1 be the unit circle in the complex plane. thus f is well de ned. f : G ! F an epimorphism and suppose that ker f = im g.42 Consider Example 3. Example 3. Then f is a well de ned epimorphism with ker f = ' kZ. Thus ' : R=2 Z ! S 1. the set de ned by jzj = 1. Note that Z2 = Zi Z(j i) = im p0 ker p0 . (b) Give an example showing that if F is not free than the conlusion may be wrong. Exercises cise 3. We de ne ' : R ! S 1 by '( ) = ei = cos + i sin .e.26 in terms of the quotient isomorphism. z = x + iy 2 C. Since. Example 3. Then ' is a homomorphism from the additive group of R to the multiplicative group S 1. (a) If F is free. f is. show that G ' H F . Then S 1 is a multiplicative group with the complex number multiplication and the unity 1 = 1 + i0.9 If m and n are relatively prime.36. If F is free.11 Let g : H ! G be a monomorphism. 3. Conclude that G0 ' F . so this is the trivial case of Theorem 3. Then q = 1G=H . i.39 Let q : G ! G=H be the canonical homomorphism from Example 3. also.2). 3.41 Let's go back to p0 in'Example 3.37. This observation will be later generalized. Example 3. show that there exists a subgroup G0 of G such that G = ker f G0. show that Zm Zn ' Zmn (see Exer- . im p0 = Zi = A and ker p0 = Z(j i) Thus f : Z2=Z(j i) ! Zi. f ((a + H ) + (b + H )) = f (a + b + H ) = f (a + b) = f (a) + f (b) it is a homomorphism.40 Let f : Z ! Zn be given by f (a) = a mod n (the remainder of a in the division by n). ' It is an epimorphism with the kernel ker ' = 2 Z. an epimorphism since im f = im f . f is a monomorphism since f (a + H ) = f (a) = 0 which is equivalent to ker f = H . 3. Thus f : Z=kZ ! Zk . HOMOMORPHISMS 93 Proof: By the preceeding discussion.10 Let f : G ! F be a homomorphism of abelian groups.3.4. Example 3. Finally.1.

g2. For a xed matrix A. Here are the three types of elementary row operations over Z : 2 6 6 6 6 4 . If f : G ! G0 is any homomorphism. then it is determined by its action on the basis elements of G. 7 6 . : : : .8) Conversely. gng and fg1. fg1. Then f : Zn ! Zm is represented by the matrix multiplication y = f (x) = Ax or. Cn its columns. denote by R1. In particular. otherwise it is not invertible. if n = m and Equation (3. In this section we discuss the analogy of this technique in the study of homomorphisms of free abelian groups. : : : . Let G and G0 be nitely generated free abelian groups with bases. 76 . more explicitely. by y1 3 2 a11 a12 a1n 3 2 x1 3 a2n 7 6 x2 7 y2 7 6 a21 a22 7 6 76 7 7 = 6 .. gmg. the columns of A are elements of the new basis for Zm expressed in terms of the canonical basis of Zm . if A = (aij ) is any n m matrix with integer coe cients. 2. respec0 0 0 tively.. Rm its rows and by C1. there are unique aij 2 Z. then the formula (3. C2. For example.5 Matrix Algebra over Z and Normal Form A basic technique in the study of linear maps of vector spaces is the row and column reduction of matrices. m. n such that CHAPTER 3.. : : : . our matrices have integer coe cients and division is not allowed.9) is a change of coordinates. we may suppose that G = Zn and G0 = Zm . 2. i = 1.9) . Thus f may be identi ed with the matrix A called the matrix of f with respect to the given bases on G and G0 . g2. ABELIAN GROUPS f (gj ) = m X i=1 aij gi0 : (3. 7 : (3.94 3. 7 6 . : : : . 7 5 4 54 5 ym am1 am2 amn xn Recall that the columns of A generate the image im A := im f . : : : . Even more. Due to the isomorphism in Example 3.34 associating any basis in G and G0 to the canonical bases in Zn and Zm . Many results of elementary matrix algebra have straightforward extensions to our case but there is one subtlety. j = 1. R2 . the operation of multiplying the i-th row of a matrix by a number a is an elementary row operation over Z if and only if a = 1. : : : ..8) extends by linearity to a unique homomorphism f : G ! G0.

where q 2 Z . If we wish to exchange the second . (c3) Replace Cj by Cj + qCl . (r1) Exchange rows Ri and Rk . in fact. and third column. (c2) Multiply Cj by 1 . (r2) Multiply Ri by 1 . The elementary column operations correspond to the right multiplication of A by elementary matrices D obtained by performing the same operation on the identity n n matrix In n. which are.43 Let A be a 5 3 matrix.3. Indeed.5. Example 3. Each operation can be expressed in terms of matrix multiplication: new matrix B is obtained by multiplying A on the left by an elementaty matrix E which is obtained by performing the same operation on the identity m m matrix Im m . MATRIX ALGEBRA OVER Z AND NORMAL FORM 95 Note that these operations are invertible over Z. (r1) and (r2) are self-inverses and the inverse of (r3) is replacing Ri by Ri qRk . row operations on the transposed matrix AT . this can be done by the elementary matrix 2 3 1 0 0 0 07 60 0 1 0 07 6 7 6 E=6 0 1 0 0 0 7 6 7 60 0 0 1 07 5 4 0 0 0 0 1 since 3 2 1 0 0 0 07 2 2 36 3 a11 a12 a13 a14 a15 6 0 0 1 0 0 7 a11 a13 a12 a14 a15 6 a21 a22 a13 a14 a15 7 6 0 1 0 0 0 7 = 6 a21 a23 a22 a24 a25 7 : 7 4 56 6 7 6 0 0 0 1 0 7 4 a31 a33 a32 a34 a35 5 a31 a32 a33 a34 a35 4 5 0 0 0 0 1 The same applies to elementary column operations over Z: (c1) Exchange columns Cj and Cl . where q 2 Z . (r3) Replace Ri by Ri + qRk .

then the equation y = Ax is equivalent to y = C x where x := E 1x. (a) The elementary row operations over Z preserve the subgroups ker A and coim A := im AT of Zn . 1) generate the row space of the initial matrix.44 Let A be an n m matrix with integer coe cients. This terminology is justi ed by the above remark that the columns of A generate im A. Proposition 3.96 CHAPTER 3. Since E is invertible. where E is an elementary matrix. if B = EA where E is an elementary matrix. Proposition 3. Example 3. 3) of Z2 generate the whole group Z2 . The following propositions are straightforward analogies of elementary linear algebra results. Let aij be the rst non-zero entry in its row Ri . De nition 3. (0. If C = AE . then the non-zero rows of A are linearly independent. . and thus they form a basis for coim A. Indeed. ABELIAN GROUPS Each row operation corresponds to a change of basis in the range space Indeed. 0). although no two of them do.45 A matrix A is in row echelon form if the following property is satis ed. Zm . where y := Ey. The group coim A is traditionally called the row space of A and im A the column space of A. (b) The elementary column operations over Z preserve the subgroups im A and coker A := ker AT of Zm .46 Suppose that A is in row echelon form. 2). Similarly.47 We show that the elements (3. then the equation y = Ax is equivalent to y = Bx. then akj = 0 for all k > j . 0) and (0. y = E 1y. (2. row operations over Z give 3 2 3 2 2 3 ! 1 2 3 2 R1 R2 1 2 6 2 0 7 ! 6 2 0 7 R2 2R1 + R3 6 0 1 7 5 4 5 5 4 4 0 3 ! 0 3 ! 0 3 3 2 R1 + 2R2 1 0 ( 1)R2 6 0 1 7 5 4 R3 3R2 0 0 hence the rst two rows (1. or x = E x. and the columns of E 1 are the new basic vectors in Zm . Thus the columns of E represent the new basic vectors in Zn . each column operation corresponds to a change of basis in the domain Zn.

Apply the row operation r1 to exchange rows 1 and k. The simultanuous row operations over Z of the identity matrix I3 3 and AT give 2 3 1 0 0 0 1 3 5 I jAT ] = 6 0 1 0 2 0 4 3 7 4 5 0 0 1 2 1 1 2 3 2 0 1 0 2 0 4 3 ! 6 1 0 0 0 1 3 5 7 = P T jC T ] . MATRIX ALGEBRA OVER Z AND NORMAL FORM 97 23 17: 7 5 17 5 3 2 We will nd bases for ker A and im A. 1.3. 7 6 . Case 1. its rst two rows (2. The rst column C1 of A has at most one nonzero entry. 3. The third row (1. 7 4 5 A0 0 Example 3. Assume that C1 has one nonzero entry ak1. 5) form a basis for im C = im A.5. 0. From now on assume m > 1. 1. Then the new matrix has the form 2a a akn 3 k1 k2 6 0 7 6 7 6 . Theorem 3. If m = 1. 5 4 1 1 1 0 0 0 0 where C = AP .49 Let A be an n m matrix with integer coe cients. 3) and (0. Since the matrix C T is in a row echelon form. The following two theorems show that the method presented in the above examples may be applied to any integer matrix.48 Let A : Z3 ! Z4 be given by 20 2 6 A=6 1 0 63 4 4 .. 4. Their proofs are constructive and may be used to obtain formal algorithms. then A = a11 a12 a13 a1n ] which is in row echelon form. Then A can be brought to a row echelon form by means of elementary row operations over Z. 1) of P T generates ker A. Proof: The proof is by induction on the number m of rows of A.

For each ai1 6= 0 apply the row operation r3 to replace Ri by Ri qiRk . Returning to the de nition of observe that (A2) = jrij < (A1): If A2 satis es Case 2. A can be reduced to row echelon form. Observe that A0 is an n 1 m 1 matrix and so by the induction arguement can be reduced using row operations to row echelon form. divides all entries of C1. Let A1 := A and let A2 be the matrix obtained by replacing Ri by Ri qiRk . A3 : : : where (A1) > (A2) > (A3) > : : : : Since any strictly decreasing sequence of positive integers is nite. then ai1 = qi + ri. If not. This results in a new rst column all of whos entries are zero except ak1. then the matrix would have a similar form even without exchanging any rows. If does not divide the entry ai1 . Without loss of generality we may assume that ak1 = . ABELIAN GROUPS If all entries in C1 were zero. there is a matrix Al which falls into Case 2. Case 3. A2. Thus. fails to divide some entry ai1 . The assumption that divides all entries of C1 is equivalent to the statement that for each ai1 6= 0 there exists qi 2 Z such that qi := ai1. Hence from now on it is assumed that C1 has multiple nonzero entries. then we are done.98 CHAPTER 3. then applying the argument of Case 3 using (A2 ) results in a matrix A3 . i 6= k of the rst column. By means of elementary row and column operations over Z. There are two cases to consider. where qi. The rst entry of the new row Ri is ri. Theorem 3. Let = (A) := minfjai1j j ai1 6= 0. 2. Thus the problem is reduced to Case 1. we could use the row operation r2 to change the sign of ak1.50 Let A 6= 0 be an n m matrix with integer coe cients. Applying Case 3 mulitple times results in a series of matrices A1. it is possible to bring . Case 2. If it does not. : : : mg Let jak1j = . ri 2 Z and O < jrij < . i = 1.

= akl fails to divide some entry of its row Rk or its column Cl . 2. As before. : : : ng: Let jaklj = .5.10) where bi are postive integers and bi divides bi+1 for all i. then A is trivially in normal form. 2. We put b1 := and use the induction hypothesis for the matrix A0 obtained by removing the rst row and rst column. 0. The following simple observation is crucial. 0]. b1 divides bi for all i > 1. : : :. : : : . then divides all entries of B . Then we apply the same arguments as in the previous proof to reduce the problem to Case 1. 0]T and using those arguments for AT gives the rst row . Case 2. Proof: The proof is essentially an a more elaborated version of the arguments of the previous proof. By the arguments of the previous proof we get a matrix whose rst column is . j = 1. . From now on assume that nm > 1. Case 1. divides all entries of A. By the above observation. There are three cases to consider. If nm = 1. MATRIX ALGEBRA OVER Z AND NORMAL FORM A to the form 99 2b 6 1 b2 6 6 6 6 6 6 0 B=6 6 6 6 6 6 6 6 4 0 bs 0 3 7 7 7 7 7 0 7 7 7. we may assume that akl = since otherwise we multiply Rk by 1. The induction is now on the totatl number of entries of the matrix nm. By row and column exchanges we get = a11 . : : : m. 0. 0. 7 7 7 7 7 7 5 0 7 (3. Observation: If an integer divides all entries of A and a matrix B is obtained from A by elementary row and column operations over Z. 0.3. Let = (A) := minfjaij jaij 6= 0 i = 1.

Then there are bases of G and G0 such that the matrix of f with respect to those bases is in the normal form (4). a basis for ker A. Then we 0 0 replace Rk by Rk = Rk + Ri0 . invertible over Z.100 CHAPTER 3.2 is called the normal form of A. We rst replace Ri by Ri qRk so to get a new i'th row Ri0 whose l'th entry is 0 and j 'th entry is aij qakj . = akl divides all entries in its row and column but it fails to divide some entry aij with i = k and j = l. In the second case. the problem is to nd one basis. 2 3 6 4 (a) A = 6 4 0 7 4 5 0 6 (b) The matrix A in Example ??. does not divide a0kj . It should be emphasized that the problem of reducing a matrix to the normal form (4) should be well distinguished from a more di cult problem of diagonalizing an n n real matrix A.51 Let f : G ! G0 be a homomorphism of nitely generated free abelian groups. 2 3 2 0 0 (c) A = 6 0 3 0 7 4 5 0 0 9 .12 For each matrix A speci ed below. nd its normal form B and two integer matrices P and Q. we reach the following Corollary 3. so the problem is reduced to Case 2. By the hypothesis. ABELIAN GROUPS Case 3. 6 6 Let q = ail 1 2 Z. such that QB = AP . Due to the relation between elementary row and column operations over Z and changes of bases discussed at the begenning of this section. The rst entry of Rk is and the j 'th entry is a0kj = (1 q)akj + aij . Exercises 3. Use the information provided by P and Q for presenting bases with respect to which the normal form is assumed. The matrix B given by Theorem 4. and a basis for im A. the same one for Rn viewed as the domain and as the range of A.

gm are of in nite order. respectively and 1 < t1 jt2 j : : : jtk The numbers m and t1 . : : : . The above theorem allows us to write G as G = F T where F= Mr i=1 Zgi T= Mk i=1 hg +ii : T is the torsion subgroup of G mentioned in Section 1 and F is a maximal free subgroup of G. : : : . : : : .52 Let G be a nitely generated abelian group.53 Let G be a nitely generated abelian group. if m. If r > 0. we get the following isomorphic to Corollary 3.9. If k = n r > 0 then gr+1. n are relatively prime. 3. DECOMPOSITION THEOREM FOR ABELIAN GROUPS 101 Theorem 3. 2.3. gm+k have nite orders t1 . tk are as in Theorem 4. The number r is the rank of F and it is called the betti number of G and the numbers t1 . Then G is Zr Z=t1 Z=t2 : : : Z=tk where r and t1 . g1 .1. : : : . gn of G and an integer 0 r n such that 1. Thus. t2 .17. : : : . t2 . 3. : : : . t2. gr+2. t2 . : : : . By Exercise 3. then Zmn ' Zm Zn . Then G can be decomposed as a direct sum of cyclic groups. tk . tk are called the torsion coe cients of G. tk are uniquely determined by G. More explicitely.6. although generators g1 .6 Decomposition Theorem for Abelian Groups The goal of this section is to prove the following decomposition theorem for nitely generated free abelian groups. g2 : : : . : : : . tk to products of primes we get the following . gn are not. there exist generators g1 . g2. t2 . g2. by decomposing the numbers t1 . By Example 3. G= Mn i=1 hgi i .

Let H Proof: Since F is a nitely generated free abelian group there is an integer n such that F = Zn. and therefore cyclic. an) 7! ai. We will use this group to de ne the above mentioned generator hm . bm . Hm is a subgroup of H and Hn = H . ABELIAN GROUPS Corollary 3. : : : . For m = 1. : : : . 0. Proposition 3. Now assume that fh1. be a subgroup of F . Let h 2 Hm. hng of elements of Zn which generate H . h2. This means that there exists km 2 Z such that < km >= m (Hm ). : : : . If m = 1. h2 . If m (Hm ) 6= 0.102 CHAPTER 3. This will be done by induction on m. Thus h = ihm + i1 h1 + i2h2 + + im 1 hm 1 . : : : . then H is nitely generated. We need to show that the set fh1 . a2 .52 requires the following results. From this it is easy to check that for all m n. then de ne hm = 0. ps are prime numbers. To show that H is nitely generated. De ne hm by m(hm ) = km . hm 1 g generates Hm 1. p2 . b2. Then m (h) = k m (hm ) for some integer j . : : : .54 Any nitely generated abelian group G is isomorphic to Zr Z=pm1 Z=pm2 : : : Z=pms 1 2 s where p1 . h2. If m(Hm ) = 0.55 Let F be a nitely generated free abelian group. then h 1(h1 )i = 1(H1 ) which implies that hh1 i = H1 or that H1 = 0. Let i : Zn ! Z be the canonical projection that sends (a1. This implies that m (h jkm ) = 0. : : : . and hence h ihm 2 Hm 1 . hng generates H . To prove Theorem 3. : : : n consider m(Hm ). : : : . 0). then m (Hm) is a nontrivial subgroup of Z. it is su cient to nd a nite collection fh1. De ne Hm := fb 2 H j i(b) = 0 if i > mg: Observe that an element of Hm is of the form (b1 . Using this isomorphism we shall identify F with Zn and think of H as a subgroup of Zn.

Then H is the row space of A. It is left as exercice to prove that tha non-zero elements of fh1. : : : m by ^ si(sj ) = 1 if j = i ^ 0 otherwise. by Proposition ?? H is a free group and r := rank H m. which form a basis for ZS . By Theorem 3. hr g for H and fz1 . thus r(H ) r(F ). Consider the free abelian group ZS . By Proposition 3. : : : . : : : . By Theorem 3. i = 1. The non-zero rows of the reduced matrix are linearly independent and hence they form a basis for H . h2 . Then by Theorem 3. hm 2 Zn generators of H . De ne f : ZS ! G by f (^i) = si.6. the number of non-zero rows of an echelon matrix is less or equal than the number n of columns. Proposition 3. : : : .49. : : : zm g for ZS such that . z2. Then any subgroup H of F is free of rank r(H ) r(F ).38. Since ZS is a nitely generated free abelian group. Consider a matrix A whose i'th row is the vector hi .55 there exist h1 .52: Let S := fs1 .56 Let F be a nitely generated free abelian group. hence they form a basis for H . Proof of Theorem 3. Using this isomorphism we shall identify F with Zn and think of H as a subgroup of Zn. h2. Recall we de ned the functions si : S ! Z.50 there exist bases fh1 . DECOMPOSITION THEOREM FOR ABELIAN GROUPS 103 and the conclusion follows. h2. sm g be a set of generators for G. Of course. Thus to prove the theorem it is su cient to obtain the desired decomposition for the group ZS =H . Let j : H ! ZS be the inclusion homomorphism. Proof: Since F is a nitely generated free abelian group there is an integer n such that F = Zn . f : ZS =H ! G is an isomorphism. A may be reduced over Z to a row echelon form.3. : : : . hng are linearly independent. This is a group homomorphism and so s H := ker f is a subgroup of ZS .

Observe that the basis for H as a subset of ZS is fb1 z1 . 6 7 60 6 7 br 7 6 7 6 7 6 7 6 7 0 4 5 where b1 1 and bi jbi+1.104 the matrix for j has the form CHAPTER 3.. : : : . br zr g. If bs+1 . 7 6 7 . then for j = s + 1. j is a monomorphism each bi 6= 0. Since. br > 1. : : : . then for i = 1. bs = 1. : : : . : : : . It is now easy to see that ZS =H = Zz1 =Zb1 z1 Zzr =Zbr zr Zzr+1 Zzi =Zbi zi = 0: Zzm : If b1 . Zzi =Zbi zi = Zbj : Therefore. b2z2 . r. s. : : : . ABELIAN GROUPS 2b 3 0 1 6 . ZS =H = Zbs+1 Zbr Zm r : .

3. called chains.11) C is a free chain complex if Cn is free for all n 2 Z. @ngn2Z consists of abelian groups Cn. with what we have learned in this Chapter we can now handle the general case.11) implies that im @n+1 ker @n and hence the following de nition makes sense. each Cn is a nitely generated free abelian group. at least in the purely algebraic setting. such that De nition 3. Recall that in Chapter 2 we were forced to deal with Homology groups in the context of vector spaces. 2. The cycles of C is the subgroup Zn := ker @n while the boundaries are the subgroups Bn := im @n+1 : Observe that (3. De nition 3.3. .59 C is a nite chain complex if: 1. and homomorphisms @n : Cn ! Cn 1.57 A chain complex C = fCn.7. there exists an N 0 such that Cn = 0 for all n > N and n < 0. We will only be concerned with free nite chain complexes in this book. called boundary op@n @n+1 = 0 (3.58 The n-th homology group of the chain complex C is Hn(C ) := cycles=boundaries = ker @n=im @n+1 : Observe that this is a purely algebraic de nition.7 Homology Groups We now turn to a purely algebraic description of Homology groups. HOMOLOGY GROUPS 105 De nition 3. erators.

w0 2 Wn. then kw 2 Bn for some k 2 Z n f0g. Bn is a group so kw 2 Bn which implies that w 2 Bn. and hence. ABELIAN GROUPS Un. then there exist nonzero integers k and k0 such that kw. Finally. and Wn of Cn such that Cn = Un Vn Wn where Theorem 3. Proof: If w 2 Wn .62 Wn Zn. Vn. k0kw. for every n 2 Z there exist subgroups 2 3 b1 07 6 @n = 6 .106 CHAPTER 3. there are bases for Un and Wn 1 for which the matrix of @n . bi j bi+1 : 4 5 0 bl Proof: Let Zn := ker @n . Since Bn is a group. Proof: 0 2 Wn since 0 2 Bn .60 (Standard Basis for Free Chain Complexes) Let C = fCn. if w. the boundaries are Bn : im @n+1 . @nw = 0. But Bn Zn hence 0 = @nkw = k@nw = 0. Therefore. Hn(C ) is a nitely generated abelian group and hence Hn(C ) = Zk Tn (C ) where Tn(C ) is the torsion subgroup of Hn(C ). k0w0 2 Bn. However. These are the cycles introduced in Chapter 2. then kw 2 Bn for some integer k 6= 0. However.61 Wn is a subgroup of Cn. . . (w + w0) 2 Wn. @n(Vn) = 0 @n(Wn) = 0: Furthermore. Lemma 3. De ne Wn := fc 2 Cn j 9 k 2 Z n f0g such that kc 2 Bng: Lemma 3. @ng be a free nite chain complex. Similarly. Consider the projection p : Hn(C ) ! Hn(C )=Tn(C ) = Zk : takes the form @n (Un) Wn 1 . 7 bi 1. Then. Wn is called the group of weak boundaries. If w 2 Wn. kk0w 2 Bn and hence k0kw + kk0w0 2 Bn which implies that k0k(w + w0) 2 Bn . Cn 1 is free.

then k c + Bn ] = kc + Bn] = Bn] = 0. : : : . 3. this standard basis can be computed using the . bl e0l g is a basis for Bn 1. On the other hand. eng is a basis for Zn. bl @n = 6 6 6 6 6 0 4 3 7 0 7 7 7 7 7 7 7 0 7 5 The following three observations follow directly from the form of this matrix: 1. Let fd1. fb1 e01.63 ker p = Wn and hence Zn=Wn = Hn(C )=Tn(C ). The proof of the theorem is nished once we de ne Un = he1 . : : : . Then Zn = Vn Wn where Vn = hc1 . : : : el i.e..64 The homology groups of a nite free chain complex C = fCn. fel+1. fe01. ck i. : : : .7. c + Bn] 2 Tn (C ).3. dlg be a basis for Wn. @ng are computable. If kc 2 Bn for some integer k 6= 0. : : : . Let fe1 . : : : . Furthermore. if for all nonzero integers k. Proof: By de nition Hn (C ) = Zn=Bn. In conclusion then c + Bn] 2 Tn(C ) if and only if c 2 Wn. Then Cn = Un Zn = Un Vn Wn where Vn and Wn are de ned as above. Proof: By the previous theorem there exists a standard basis for the free chain complex. : : : . HOMOLOGY GROUPS 107 Lemma 3. then k c + Bn] 6= 0 for all k 2 Z n f0g. 2. ck g be a basis for Zn=Wn. : : : . Let fc1 . kc 62 Bn. ej g be a basis for Cn and let fe01 . e0l g is a basis for Wn 1 . Thus. e0mg be a basis for Cn 1 such that @n : Cn ! Cn 1 has the form 2b 0 6 1 . Theorem 3. So cosets in Hn (C ) have the form c + Bn. 6 6 6 0 . h c + Bn]i = Z. i. : : : .

Vn and Wn. The relative homology groups are Hn((C . ABELIAN GROUPS row and column reductions described in Theorem 3. In this basis we can identify the subgroups Un . Furthermore.65 Let C = fCn.50. @n g is a subchain complex of C if: 1. The relative n-boundaries are Bn(C .66 The relative n-cycles are Zn(C . complex is just the boundary operator of the larger complex restricted in its domain. @ng be a chain complex. 0 2. D): .108 CHAPTER 3. @n = @n jDn . A chain complex 0 D = fDn. We can create a new chain complex called the relative chain complex whose chains consist of the groups Cn =Dn and whose boundary operators are the induced maps @n : Cn=Dn ! Cn+1=Dn+1 given by c + Dn] 7! @n c + Dn 1]: @n is well de ned since @n(Dn) Dn 1. D) := ker @n+1. D) := Zn(C . For this reason and to simplify the notation we shall let @ 0 = @ . Bn = im Un+1 and Hn(C ) = Vn Wn=Bn: Before ending this section we will introduce yet another construction that leads to homology groups. 0 The condition that @n = @n jDn indicates the boundary operator of a subchain De nition 3. @n g be a chain complex and let D = fDn. D) := ker @n . @ng be a subchain complex. Dn is a subgroup of Cn for all n 2 Z. 0 Let C = fCn. @n @n+1 c + Dn+1 ] = @n @n+1 c + Dn] = @n @n+1 c + Dn 1] = 0 + Dn 1] = 0: De nition 3. D)=Bn(C .

2 a chain of small squares called in computer graphics pixels. But that seems to be very arti cial: what we see does already have a nice combinatoric structure and we should be able to extract algebra out of it. There are several ways to extract combinatoric and algebra information from a set in Rn. An approach arising naturally from numerical computations and graphics is by means of cubical grids which subdivide the space to cubes with vertices in an integer lattice. Note that any two pixels are either disjoint. The classical approach is by means of triangulations of the space. In particular. there is only a nite amount of information involved. In this chapter we shall formally de ned cubical homology. For example if n = 2 that means subdividing the space into triangles so that any two triangles are either disjoint. intersect at a common edge. the rst step is to generalize the combinatorics of graphs to higher dimensional spaces. 109 . But. empasizing strong and weak points of each approach. The classical Simplicial Homology Theory would require from us subdividing each pixel to a union of at least two triangles in order to compute homology. Look for example at Figure 4. If we blow up a section of the gure we will see in Figure 4. The algebra of triangulations is the Simplicial Homology Theory. This approach is the Cubical Homology Theory presented here. At the end of this chapter we shall give a brief overview of the Simplicial Homology and compare the two theories.Chapter 4 Cubical Homology In Sections ?? we suggested what were the important elements in Homology. However.1 The picture seems to be composed of curves which do not look like polygonal curves. like any picture produced by a computer. or at a vertex. intersect at a common edge or at a vertex. we used the edges and vertices of a graph to generate algebraic objects that measured the nontriviality of the topology of the graph.

Similarly. Later on we will investigate what happens with the algebra extracted from a cubical grid when we change units. To simplify the notation we will use the notation l] = l. De nition 4. 3] is not an elementary interval since the length of the interval is greater than 1.1 An elementary interval is a closed interval I R of the I = l. 1 . 14]. and 7] are all examples of elementary intervals. Q = I1 I2 In Rn .1 Cubical Sets form 4. it has sides of length 1 and vertices with integer coordinates. 1. Elementary intervals that consist of a single point are degenerate. On the other hand.e.110 CHAPTER 4. l + 1] or I = l.2 The intervals 2.1 Elementary Cubes De nition 4.2: A blow up of the previous gure. 3 ] is not an elementary inter2 2 val since the boundary points are not integers. 4. In numerical and graphical analysis one needs to consider very ne cubical grids.e.1. With appropriate units we may assume in this chapter that each cube is unitary i.1: A typical computer graphics picture. Elementary intervals of length one are nondegenerate. i. From a theoretical point of view. The size of cubes of a grid cannot be arbitrarily small because of the computer's capacity.3 An elementary cube Q is a nite product of elementary intervals. l] for some k 2 Z. l] for an interval that contains only one point. CUBICAL HOMOLOGY Figure 4. Figure 4. 15. 3]. Example 4. the size of a grid is just a question of choice of units.

1.g. Thus. However.4. we will take great care in this book to explicitly state this identi cation if we make it. 2] R The elementary cubes 1. 2] R2 3 2 1 1 1 3 2 1 1 t t t t 1 2 3 4 1 1 2 3 4 Figure 4. i. 2] 1] R2 The elementary cube 1. 1] 2. The set of all elementary cubes in Rn is denoted by Kn . Of course there are many other elementary cubes. K := Figure 4. 2] 0. Of course using the inclusion map : R ! R2 given by (x) = (x.e. Observe that the cube 1. 2] 1] R2 and 1. CUBICAL SETS 1 111 where each Ii is an elementary interval. if the identi cation is not clearly stated. Q1 := 1. 1] R3 .3: Elementary cubes in R and R2. The set of all elementary cubes is denoted by K. then they should be treated as distinct sets.3 indicates a variety of elementary cubes. 0) we can identify these two elementary cubes. 2] 0] R2 since they are subsets of di erent spaces. 2] R is di erent from the cube 1. e. 2] 1. t t 1 0 1 2 3 4 n=1 Kn : The elementary cube 1.

The embedding number of Q is denoted by emb Q and is de ned to be n since Q Rn. Using this notation we can write Kn := fQ 2 K j emb Q = ng: Similarly. De nition 4. 0) 2 R3 := 1. The dimension of Q is denoted by dim Q and is de ned to be the number of nondegenerate intervals Ii which are used to de ne Q. 4] f6g Q2 Q3 Q4 Q5 1. 1] R3 := 1. 1] = f1g 1. we will let Kd := fQ 2 K j dim Q = dg and n Kd := Kd \ Kn: Example 4. then n+m Q P 2 Kd+k : (4. CUBICAL HOMOLOGY := 1] 1.4 Let Q = I1 I2 In Rn be an elementary cube.5 Refering to the elementary cubes de ned above we have that emb Q1 = 3 emb Q2 = 3 emb Q3 = 3 emb Q4 = 3 emb Q5 = 4 and and and and and dim Q1 = 3 dim Q2 = 2 dim Q3 = 1 dim Q4 = 0 dim Q5 = 3 In particular. 2] 0] 1] = 1. 2] 0.112 CHAPTER 4. 0] 3. 2] f0g f 1g R3 := 0] 0] 0] = (0.6 Let Q 2 Kd and P 2 Kk . the reader should observe that the only general relation between the embedding number and the dimension of an elementary cube Q is that 0 dim Q emb Q: n m Proposition 4. 0. 0] 3. 2] 0.1) . 2] R4 which we shall not attempt to draw. 2] = 1. 4] 6] 1.

we can write P = J1 J2 : : : Jm where each Ji is an elementary interval.1. 1].6 that though they lie in the same space Q P 6= P Q. l We extend this de nition to a general elementary cube Q = I1 I2 : : : In Rn by de ning the associated open elementary cube as Q := I 1 I 2 : : : I n: .2 Representable Sets De nition 4.1. ] if I = l. Hence. Exercises 4. CUBICAL SETS 113 Proof: Since Q 2 Kn it can be written as the product of n elementary intervals. It should clear from the proof of Proposition 4.e.7 Let I be an elementary interval. Elementary cubes will be the building blocks for the homology theory that we will develop.4.1 Prove that any elementary cube is closed. l]. It is left to the reader to check that dim(Q P ) = dim Q + dim P . For this reason we introduce the notion of open cubes. Q P = I1 I2 : : : In J1 J2 : : : Jm which is a product of elementary intervals. Q = I1 I2 : : : In: Similarly. i. l + 1) if I = l. The associated open elementary interval is + I := (ll. 4. however for technical reasons it will useful to have additional sets to work with.

g < 1. Proof: Since Q is an elementary cube it is the product of elementary intervals Q = I1 I2 In Rn. Proposition 4. bi] where ai 2 Z and bi = ai or bi = ai + 1. ) 6 Q.4). (iii) If P.8 An important word of warning: An open cube need not be an 1 open set. The associated open elementary cube Q is an n open set if and only if Q 2 Kn for n 1. 2 n Assume that Q 2 Kd where d < n. 2 Consider now the elementary cube Q = 1.114 CHAPTER 4. (ii) A Rn bounded implies that card fQ 2 Kn j Q \ A 6= . then by Exercise 1. Therefore.10 We have the following properties (i) Rn = fQ j Q 2 Kn g. We can generalize this example to the following Proposition. Observe that x = (x1 . xn) 2 Q. Proposition 4. (v) Q 2 Kn implies that Q = fP j P 2 Kn P S S Qg. if Q 2 Kn . then the open elementary interval I R is an open set. 2) 3] R2 which is clearly not an open set. . (iv) For every Q 2 K. x2.9 Let Q 2 K. This is a single point and hence a closed set (see Exercise 1. The associated open elementary cube is Q = (1.4 Q is an open set. Of course. CUBICAL HOMOLOGY Example 4. Thus. Q is not open. if I 2 K1 . Then. Consider for example 1] 2 K0 . cl Q = Q. Q 2 Kn .4 any 1 interval of the form (l. Let xi = ai +bi . 2] 3] 2 Kd. Let Ii = ai . : : : . or P = Q. as was shown in Exercise 1. there exists i0 such that Ii0 = ai0 ] is a degenerate interval. B (x. Observe that for any > 0. then P \ Q = . l + 1) is an open subset of R. n On the other hand.

B := fi = 1. : : : .10(v) we get . xi] if xi is an endpoint of Ii otherwise. This proves (i). xn) 2 Rn and put i. To prove the opposite inclusion take an x = (x1 . it extends immediately to elementary cubes of dimension greater than one. : : : . It follows that x 2 cl Q. : : : . (iv) Observe that Q Q. l2 ] : : : kn. !1 (v) Consider Q = I1 I2 : : : In and let x = (x1 . n j xi = lig: j j j De ne yj := (y1 . Hence x belongs to the right-hand-side of (v). (ii) The proof is straightforward. i and put P := J1 J2 : : : Jn Then obviously x 2 P and P Q. De ne Ji := Ixi . De nition 4. To prove the opposite inclusion take an x = (x1. Using open cubes we can de ne a class of topological spaces. < j := x i 2 A \ B or i 62 A B . i i Then Q := I 1 I 2 : : : I n is an open cube and x 2 Q. The family of representable sets in Rn is denoted by Rn. : : : . : : : yd) 2 Rn by 8 > xi + 21n i 2 A n B . because the intersection of Cartesian products of intervals is the Cartesian product of the intersections of the corresponding intervals. CUBICAL SETS S 115 Proof: (i) Obviously fQ j Q 2 Kng Rn. xn) 2 Q. yi > i : xi 1 i 2 B n A. y2. ln] and put A := fi = 1.1. n j xi = kig. Let Q = k1. x2. x2 : : : xn ) 2 Q. Also. otherwise. x2 . oor (x ) + 1] if xi 2 Z. (iii) For elementary cubes of dimesion one the result is obvious. 2n Then yj 2 Q and jlim yj = x.4. therefore cl Q Q.11 A set Y Rn is representable if it is a nite union of open elementary cubes. As an immediate consequence of Proposition 4. l1] k2 . x ] Ii := xoori(x ).

(iii) The set ch (A) is closed and representable. (vi) oh (oh (A)) = oh (A) and ch (ch (A)) = ch (A).3) (4.14 Consider the vertex P = 0] 0] 2 R2. Proposition 4. and the closed hull of A is ch (A) := fQ j Q 2 K. In particular. De nition 4. (ii) The set oh (A) is open and representable.2) Example 4. Q \ A 6= . then ch (x) ch (y). Proposition 4. (iv) oh (A) = TfU 2 Rn j U is open and A U g (v) ch (A) = TfB 2 Rn j B is closed and A B g. an + 1): oh (P ) = (a1 1.g.16 Assume A Rn. (vii) If y 2 oh (x).116 CHAPTER 4. oh (P ) = f(x1. (viii) Q 2 Kn and x 2Q implies that ch (x) = Q: . a1 + 1) The names chosen for oh (A) and ch (A) are justi ed by the following proposition. (an 1. then ch (A) K . Then.13 The open hull of a set A Rn is oh (A) := fQj Q 2 K.15 Let P = a1 ] Then. an ] 2 Rn be an elementary vertex. if K is a cubical set such that A K .g: (4. CUBICAL HOMOLOGY Proposition 4. x2 ) 2 R2 j 1 < xi < 1g: Generalizing this example leads to the following result. Q \A 6= .12 Every elementary cube is representable. Then (i) A oh (A) and A ch (A).

Therefore Q = cl Q P . P \ A 6= .10(ii) the union in (4. Proof: (i) That A ch (A) follows directly from the de nition and A oh (A) follows from Proposition 4. Hence Q K and x 2 K . oh (x) = oh (y) and ch (x) = ch (y). Let x 2 oh (A). Q \U = . a contradiction. i. and consequently Q K . (iv) Observe that since oh (A) is open. This shows that ch (A) K and since K is arbirtary. and x 2Q. Therefore the set oh (A) is representable.. Take a z 2 ch (x).4. (iii) The set ch (A) is closed since it is the nite union of closed sets. = Q \ U = cl Q \U .e. We will show that ch (A) K . \ \ fB 2 Rn j B is closed and A B g ch (A): Let K 2 Rn be a closed set which contains A. This means that P oh (A). y 2Q.1. Assume that P \ oh (A) 6= . Since P is representable. Let P 2 Kd be such that P \oh (A) = . representable and contains A. i. (vii) Observe that since y 2 oh (x). hence x 2 U . Then there exists a Q 2 K such that Q \A 6= .e. \ oh (A) fU 2 Rn j U is open and A U g: (v) Since ch (A) is closed. For this end take an x 2 ch (A). Then there exists a Q 2 K such that Q \ A 6= . 6 It follows that . there exists a P 2 K such that y 2P and x 2 P . Then. This shows that oh (A) U and since U is arbirtary. It follows that oh (A) is open..10(v). representable and contains A.2) is nite. CUBICAL SETS 117 (ix) Let Q 2 Kn and let x. By Proposition 4.. Then there exists a Q 2 K such that z 2 Q . To prove that oh (A) is open we will show that it satis es ??. By Proposition ?? 6 6 Q U . and x 2 Q.12 ch (A) is representable. fU 2 Rn j U is open and A U g oh (A): To show the opposite inclusion take an open set U 2 Rn such that A U .. it follows from Proposition ?? that Q P . (ii) By Proposition 4. \ ch (A) fB 2 Rn j B is closed and A B g: (vi) This follows immediately from (iv) and (v). Then there exists a Q 2 K such that A \ Q = . and P \ Q6= . It follows that Q \K 6= ..

This is an K3(X ) = 0. 1] 0. the elements of K0(X ) are the vertices of X and the elements of K1 (X ) are the edges of X . 1] 0.18 Consider the set X = 0. 1] 0] 0. 4. since X Rn. (ix) Let z 2 oh (x). which proves (vii). is a cubical set. As was mentioned before elementary cubes will make up the basic building blocks for our homology theory. CUBICAL HOMOLOGY and x 2Q. The equality ch (x) = ch (y) follows from (viii). 1] 0]. Example 4. If X and De nition 4. 1] 0. It follows that P Q. 0. 1]. Therefore. and hence.17 A set X Rn is cubical if X can be written as a nite Rn is a cubical set. when it n is convenient we will write Kk (X ). 1].3 Cubical Sets union of elementary cubes. 1] K2(X ) = f 0] 0. 1] 1]g . hence also Q P and consequently z 2 P . but it serves to reminds us that X Rn. the elements of Kk (X ) are the k-cubes of X . 1] R3. The same way one proves that oh (y) oh (x). This leads to the following de nition. 1] 0. This in turn implies that Q 2 Kn so to use the notation Kn(X ) is somewhat redundant. (viii) This is straightforward.1. Consequently z 2 oh (y) and oh (x) oh (y). It follows that Q P . 1] 0. It is easy to check that Kk (X ) := fQ 2 K(X ) j dim Q = kg: Observe that if Q X and Q 2 K then emb Q = n. 1] 0. 1] 0. 0. More generally. In analogy with graphs. 1] 1] 0. i. 1] 0. 1].e. y 2 Q. 1]. 1] 0. K(X ) := fQ 2 K j Q X g elementary cube. 0. then we shall adopt the following notation.118 CHAPTER 4. Then there exists a Q 2 K such that z 2Q and x 2 Q. 0.

4. even a simple set such as a point may or may not be a cubical set. 1]. 1] 1] 0. Kmax(X ) is the set of maximal faces in X . extremely restrictive. A face which is a proper face of exactly one elementary cube is a free face. 1] 0. Then X B (0. 0] 0. and z are all integers. Example 4. then it is a maximal face. denoted by Q X . K0(X ) K1(X ) K2(X ) is the set of proper faces.:::nfjlij + 1g max Now set R = maxQ2K(X ) (Q). 0] 0] 1]. By Exercise 4. 1] 1]. The set of free faces is given by K2(X ). In particular consider the point P = (x. 0.21 Any Q 2 K(X ) is called a face of X and is denoted by Q X . For example. 1] 0]. Example 4. let Q 2 K(X ) then Q = I1 I2 In where Ii = li ] or Ii = li. 1] 0] 0. li + 1]. . 0] 1] 0. 1] 0] 1]. 1] 0] 0]. 1] 1]. P is a cubical set if and only if x. y.1. 0.15 the nite union of closed sets is closed. Then. 1] 0. 1] 0. To show that X is bounded.20 If X Rn is cubical. 0. y. 1]. then X is closed and bounded. 1] 0. Proof: By de nition a cubical set is the nite union of elementary cubes. 0] 1] 1]. 1] 1] 1]g K0 (X ) = f 0] 0] 0]. CUBICAL SETS 0] 0. 1]. the unit circle x2 + y2 = 1 is not a cubical set. 0. R).1 an elementary cube is closed and by Theorem 1. In fact. 1] 1] 0]. Q is a proper face in X .22 Let X = 0. 1]. 1]. 1] 0] 1]. if there exists P 2 K(X ) such that P 6= Q and Q K(P ).19 It should be noted that the de nition of a cubical set is Proposition 4. If Q is not a proper face. Let (Q) = i=1. 1] 1] 0]. 1] 1] 1]g: 119 K1 (X ) = f 0] 0] 0. 1] 0]. z) 2 R3. De nition 4. 0] 1] 0]. 1] 0] 0].

23 Refering to the cubical set X The following elementary cubes are free faces 1] 2] 0.5 0 0. 3 R2 shown in Figure 4.1] -0. CUBICAL HOMOLOGY Example 4. Let X Rn consist of a single point. f is a homeomorphism and f (X ) = 0 is a cubical set.1] x [0. 1] 0]. the set consisting of a point can be represented by a cubical set as follows.2] [0. 0] 2.120 CHAPTER 4.0] x [2] 2 [-1] x [2] [0] x [2] 1.5 [0] x [1.5 -1 -0. Then.1] x [0] [0] x [0.4: Elementary cubes of X Exercises R2 .5 [-1.e.5 2 Figure 4.5 1 1. X = fx0g. 1] 1]. Let f : X ! 0 2 Rn. 1] 0.1] [1] x [0. 4. set. 0. 1]. However.1] [0.5 -1 -2 -1.5 [0] x [0. 1] 0.2 In Example 4.19 it was noted that a given point need not be a cubical .1] 0 [0] x [0] [0. i.4. Prove that any abstract graph which is a tree can be represented as a cubical set.1] x [1] 1 [0] x [1] [1] x [1] 0.

4. CUBICAL SETS 121 dimension 0 or 1 is a graph and hence gives rise to an abstract graph.4.1.3 Observe that any cubical set which consists of elementary cubes of . Give an example of an abstract graph which which does not arise as a cubical set.

Ck (X ) is the nitely generated free abelian group generated by the elements of Kk (X ) and is refered to as the set of k-chains of X . Furthermore.25 Let X Rn be a cubical set. In practice we will be interested in the chains generated by cubical sets. for k < 0 and k > n.122 4. and similarly. Ck := ZKk : If c 2 Ck then dim c := k. i. Observe that Ck (X ) is a subgroup of Ck . CUBICAL HOMOLOGY In this section we nally present the formal de nitions that we use to transition between the topology of a cubical set and the algebra of homology theory. De nition 4. Since Kk (X ) = .24 The group Ck of k-dimensional chains (k-chains for short) of X is the free abelian group generated by elements of Kk .4) 0 otherwise. b Given an elementary cube Q we will refer to Q as its dual elementary b chain. Since X is a cubical set and Kk (X ) is a basis for Ck (X ). De nition 4. Ck (X ) is nite dimensional. given any c 2 Ck (X ) there are integers ai such that X b c= aiQi : b b Qi2Kk (X ) . We begin by de ning the algebraic objects of interest. the corresponding group of k-chains is Ck (X ) = 0.1 Cubical Chains CHAPTER 4. given an elementary chain Q we will refer to Q as its dual elementary cube. b b bn Let Kk (X ) := fQ j Q 2 Kk (X )g. Observe that Ck is an in nitely generated free abelian group.2.2 The Algebra of Cubical Sets 4. Recall from de nition given in Chapter 3 that this implies that the basis b for Ck (X ) is the set of functions Q : Kk (X ) ! Z de ned by b Q(P ) = 1 if P = Q (4.e.

Proof: (i) By de nition the 0 chain is the element of the free abelian group which is not generated by any cube. l. (ii) This follows directly from the de nition of support and (i). = jc cj = jcj jcj = jcj = . (iii) This too follows directly from the dePnition of chains and support. j if a = 0 6 b (iii) If Q 2 K. : : : . The support of the chain c is the cubical set 123 b De nition 4.28 It is not true in general that jc1 + c2j = jc1j jc2j. (iv) jc1 + c2 j jc1j jc2 j. Then c1 + c2 = m l X b X b ai Qi + bj Pi: i=1 j =1 Thus. Thus.26 Let c 2 Ck (X ) and let c = Pm aiQi where ai 6= 0 for i=1 jcj := Qi Rn: i=1 m Proposition 4. Consider any chain c such that jcj = . its support jcj is a set.4.: 6 6 Notice that while a chain c is an algebraic object. m and j = 1.. x 2 jc1 + c2j implies x 2 jc1j or x 2 jc2 j. : : : .2. then jQj = Q. THE ALGEBRA OF CUBICAL SETS i = 1. m.. (ii) Let a 2 Z. we have just de ned a way to go from a cubical set to a nite dimensional free group. Observe that 6 . bj 6= 0 for i=1 i = 1. : : : . b b (iv) Let c1 = Pm ai Qi and let c2 = lj=1 bj Pi where ai. Example 4.27 Support has the following properties: (i) j0j = . . then jacj = ..cj if a = 0. and from an element of the free group back to a cubical set.

j0j = 0 2 R. Q 2 Kk and P = Q. since 2] c Q) = 1 if Q = 2] 2]( 0 otherwise. 0 2 Ck (X ) is the identity element of the group and hence j0j = . Let X R be a cubical set.e. B1 = 0. This is di erent from c 2 C0 (X ). A c represents a counter-clockwise closed path around the square. With this in mind. In particular j c j = j2 c j = 1 2 R while j c j = 2 2 R. b Proof: Since Kk is de ned to be the image of it is obvious that is b b surjective. 1] 1] 2] Finally. 1]2 shown on Figure 4. CUBICAL HOMOLOGY b b Proposition 4. 1]. In particular. A2 = 1] 0.5. Thus. Consider c 2 C0 (X ). i. 0) to (0.124 CHAPTER 4. b we think of A1 as indicating moving along the edge from (0. positive or negative elementary chains represent the direction in which an edge is traversed. 1) while b1 suggests covering the edge in the opposite direction. This implies that and hence that P = Q. 1] 1] Then jcj is the contour of the square 0. in Figure 4. B2 = 0.5 we included an orientation to the edges indicated by the arrows. .30 While the notation we are using for chains is consistent with b b b b Example 4. b b 1 = P (P ) = Q(P ) that of earlier chapters some care must be taken when discussing 0-chains that are generated by elementary cubes in R. For example. By de nition it is the function 1] c Q) = 1 if Q = 1] 1]( 0 otherwise.31 Let c = A2 A1 + B1 B2. where A1 = 0] 0. 1]. 0] Remark 4. b while c is the dual of the vertex located at the origin.. while 2 c Q) = 2 if Q = 1] 1]( 0 otherwise.29 The map : Kk ! Kk given by (Q) = Q is a bijection. To prove injectivity assume that P. 1] 0]. In addition we have chosen to give a geometric interpretion of the signs appearing in the expression for c.

ai 6= 0: i=1 Ck (K L) = Ck (K ) + Ck (L): b Let A P fi j Qi K g and B := f1. 1). : : : . Put c1 := Pi2A aiQi . In particular Qi \K = .2. 0) ? b A2 - b B1 t (1.33 If K. In terms of the basis elements this can be written as c= m X b aiQi. The b b same argument applies to Q 2 Kk (L) and so Ck (K ) + Ck (L) Ck (K L). mg n A. 2. 1) 6 b A1 t (0. 125 b B2 Q t(1. Then Q 2 Kk (K ) and hence Q 2 Kk (K L). 0) Figure 4. Consequently Qi L and since L is closed also Qi L. the chain Example 4. To prove the opposite inclusion let c 2 Ck (K L).. consider the chain 2c. Proposition 4. THE ALGEBRA OF CUBICAL SETS (0t 1) . Hence jc2j L. object. L Rn are cubical sets. The chain 2c can be interpreted as a path winding twice around the square in the counter-clockwise direction. then b b Proof: Let Q 2 Kk (K ).32 With the notation of the previous example. Then Qi K L and Qi 6 K . It follows that c = c1 + c2 2 Ck (K ) + Ck (L). 0) to (1. Obviously jc1j K . Let i 2 B .5: Boundary of the unit square. . := b c2 := i2B aiQi . Similarly.4. It is clear that j2cj = jcj so both chains represent the same geometric b b b b b b b b A1 + A2 + B1 + B2 = A1 + B2 + A2 + B1 could be interpreted as a \sum" of two di erent paths along the boundary of the square connecting (0.

De ne X c1 c2 := aibj Pi dQj : i. 1]. The cubical product has the following properties.6 indicates the support of the chains c1 . Example 4. Then (i) c1 0 = 0 c1 = 0 . By de nition we have c1 c2 = P1 dQ1 + P2 dQ1 + P3 dQ1 + P4 dQ1 + P1 dQ2 + P2 dQ2 + P3 dQ2 + P4 dQ2 while c2 c1 = Q1d P1 + Q1d P2 + Q1d P3 + Q1d P4 + Q2d P1 + Q2d P2 + Q2d P3 + Q2d P4 Figure 4. 1]. c1 c2 and c2 c1. 1]. P3 = 0. 1] 1] b b b b then Pi 2 K1 .j De nition 4. Let c1 2 Kk b and let c2 2 Kk0 . then Qi 2 K1. This motivates the following de nition. CUBICAL HOMOLOGY From Proposition 4. c3 be chains. P2 = 1] 0. b b P Q := P dQ: b We can extend this product to : Ck Ck0 ! Ck+k0 as follows. Proposition 4.36 Let c1.6 we know that the product of two elementary cubes is again an elementary cube. P4 = 0. This gives rise to chains c1 = P1 + P2 + P3 + P4 and c2 = Q1 + Q2 .126 CHAPTER 4.34 Given two elementary cubes P 2 Kk and Q 2 Kk0 set The element c c2 2 Ck+k0 is called the cubical product of c1 and c2. c2 . By de nition we can write X c X c c1 = ai Pi and c2 = bj Qj where fPig = Kk and fQj g = Kk0 . 1] 0]. 0] and Q2 = 0. c2. Let Q1 = 1.35 Let P1 = 0] 0.

2. ai bj = 0 for any i = 1. : : : . .5 1 0. then c1 = 0 or c2 = 0. k. THE ALGEBRA OF CUBICAL SETS Support of chain c 1 2 1.2. b b (iv) Assume that c1 = Pk=1 ai Pi and c2 = Plj=1 bj Qj . It follows that ( a2 )( i b2 ) = j i=1 j =1 (ai bj )2 = 0.6: The support of the chains c. (iii) The proof is straightforward. c c0 and c0 c. j = 1. : : : . 2.5 -1 -1 0 1 2 Support of chain c ♦ c 2 127 Support of chain c 2 P4 P 1 P2 P3 -1 Q 1 0 Q 2 1 Support of chain c ♦ c 1 2 1 2 2 1 0 0 -2 2 1 0 -1 -1 0 2 1 -1 2 1 0 -1 -2 0 2 Figure 4.4.5 0 -0. l.e. Proof: (i) and (ii) follow immediately from the de nition. (ii) c1 (c2 + c3 ) = c1 c2 + c1 c3 (iii) (c1 c2 ) c3 = c1 (c2 c3 ) (iv) if c1 c2 = 0. Then i k l XX i=1 j =1 k X i=1 l X j =1 b b ai bj Pi Qj = 0. k l XX i. c0.

At times the notation @k is too cumbersome. Q = I1 I2 In: b b b Set I = I1 and P := I2 I3 In. Since I1.e.2.e.37 Let Q be an elementary cubical chain such that emb Q > b b 1. then Q = I P .2 The Boundary Operator De nition 4. there exist unique elementary cubical chains I and P with emb I = 1 and emb P = d 1 such that b b b Q = I P: b Proof: Since Q is an elementary cubical chain. Notice that if @ is a linear map then it must be the case that @ 0 := 0: . 4. To obtain a free chain complex we n need to de ne boundary operators. J R. it follows that I1 = J and P = P 0. i i j b Proposition 4. Then. the chains Ck (X ) are the free groups which will be used to de ne the homology groups.38 The cubical boundary operator @k : Ck ! Ck 1 is de ned by induction on the embedding number. CUBICAL HOMOLOGY hence Pk=1 a2 = 0 or Plj=1 b2 = 0. b b c We still need to prove that this is the unique decomposition. Q is an elementary cube. i.29 we obtain I1 P = J P 0. Consequently c1 = 0 or c2 = 0. Given a cubical set X Rn. i. Since @k is supposed to be linear and Ck (X ) is a free group it is su cient to give the de nition in terms of the basis elements of Ck (X ).128 CHAPTER 4. linear maps @k : Ck (X ) ! Ck 1(X ) with the property that @k @k+1 = 0. so we will typically simplify it to @ . If Q = J P 0 for some J 2 K1 and P 0 2 Kn 1 then I1dP = J dP 0 and from Proposition 4.

ld 1] c] l0 .l+1] l. Then. ld 1] l. b @Q = = = = d + 0+ 0+ @ l.2. l + 1] 2 K1 for some l 2 Z. This matches our intuitive notions developed for graphs. THE ALGEBRA OF CUBICAL SETS ( 129 1 b b1 Let Q 2 Kk . ld 1] ( l0d1] c]) + l + + l0 0+ 0+ l d1] l0. P with emb I = 1 and emb P = n 1 such that b b b Q = I P: De ne b b b b b @ Q := @ I P + ( 1)dim I I @ P: b Finally. i. l +d l0 ] l.4. l0 + 1] l] dl0 + 1] + l. ld 1] l0. l0 + 1]. ld 1] @ l0 . Example 4.41 Let c and c0 be cubical chains. ld 1] l0d1] + l. ld 1] + ( 1)dim l. ld 1] l. b bn Now assume that Q 2 Kk where n > 1. By Proposition 4.e. we extend the de nition to all chains by linearity. b @Q = = = l l @ c] c] + ( 1)dim b] c] @ c] l l0 l0 0 c] + c] 0 l0 l 0 + 0: Thus. Then. the boundary of the dual to a vertex is trivial. l + 1] l0 .39 Let Q = l] l0 ]. then .l + 1]. then Q is an elementary interval and hence Q = l] 2 K0 or 1 Q = l. ld 1] + 0+ ( l d1] c]) l0. ld 1] c] + l + + + l0 l + 1] d0 .37 there exist b b unique elementary cubical chains I.40 Let Q = l. + l if Q = l. De ne b @ Q := 0l d1] c] if Q = l]. l + 1]d l0 + 1]: l l0 . 1] @ (c c0 ) = @c c0 + ( 1)dim cc @c0 : Proposition 4. if c = a1 Q1 + b 2 + + am Qm then b a2 Q b b b @c := a1 @ Q1 + a2@ Q2 + + am @ Qm : Example 4.

Then. Q = I P where emb I = 1 and emb P = n 1. If n = 1. i.e.37. b c b b c @ (Q Q0) = @ (I P Q0) b b c b b c = @ I P Q0 + ( 1)dim I I @ (P Q0 ) b b c b b c b c = @ I P Q0 + ( 1)dim I I @ P Q0 + ( 1)dim P P @ Q0 b b c b b c b b c = @ I P Q0 + ( 1)dim I I @ P Q0 + ( 1)dim I +dim P I P @ Q0 b c b b c b b = @ I P + ( 1)dim I I @ P Q0 + ( 1)dim QQ @ Q0 b c b c = @ Q Q0 + ( 1)dim QQ @ Q0 b b b Corollary 4.e.42 If Q1 . . Q2 . to show that b Qj 1 b b @ Qj Qj+1 b Qm : As was indicated earlier we are really interested in @k : Ck (X ) ! Ck 1(X ) where X is a cubical set. then the result follows from calculations similar to those of Example 4. Before we can employ this de nition we need to be sure that @k (Ck (X )) Ck 1(X ). then b b @ (Q1 Q2 X Pj 1 b b Qm) = ( 1) i=1 dim Qi Q1 m j =1 Proof: Since @ is a linear operator it is su cient to prove the proposition for elementary cubical chains. be De nition 4. : : : . Observe that since @ is a linear operator the following proposition su ces. Assume that it is Q that can be decomposed.40. Qm are elementary cubical chains. i. If n > 1. then we can decompose Q or Q0 as in Proposition 4.43 The boundary operator for the cubical set X is de ned to @k : Ck (X ) ! Ck 1(X ) obtained by restricting @ : Ck ! Ck 1 to Ck (X ). CUBICAL HOMOLOGY b c b c b c @ (Q Q0) = @ Q Q0 + ( 1)dim QQ @ Q0 : The proof will be done by induction on the embedding dimension of the corresponding cubes.130 CHAPTER 4.

then @k : Ck (Q) ! Ck 1(Q): Proof: Let Q = I1 I2 m j =1 b X Pj 1 b @ (Q) = ( 1) i=1 dim Ii I1 b I1 b Ij In. Again. This implies that b I1 b Ij 1 b b @ Ij Ij+1 b Im = = b I1 b I1 b I1 c]) b lj Im b Im + b Im Both terms on the right side are in Ck 1(Q) since I1 and Ij Ij 1 1 lj ] Ij+1 lj + 1] Ij+1 In Q In Q: I1 The following proposition shows that @ is a boundary operator. then Ij = lj . . lj + 1].45 @ @=0 Proof: Because @ is a linear operator it is enough to verify this property for elementary cubical chains. then 1 b b @ Ij Ij+1 b Im = 0 2 Ck 1(Q): ( ljd 1] + ljd 1] + c] lj On the other hand if Ij is nondegenerate. By Corollary 4. Proposition 4.4. THE ALGEBRA OF CUBICAL SETS 131 Proposition 4.44 Let Q Rn be an elementary cube.42 b Ij 1 b b @ Ij Ij+1 b Im: Consider each term of this sum separately. If Ij is a degenerate interval.2. the proof is by induction on the embedding number.

b then dim @ I = 0 and hence the two terms cancel each other giving the desired result. if dim I = 1. b b Observe that if dim I = 0.37 we can write Q = I P where emb I = 1 and emb P = n 1. then by de nition @ Q = 0 so b @ (@ Q) = 0.2. Then by Proposition 4.132 CHAPTER 4. If Q = l. De nition 4. 4.3 Homology of Cubical Sets Let X Rn be a cubical set. Thus we can make the following de nition. If Q = l]. CUBICAL HOMOLOGY b Let Q be an elementary interval. Then K(X ) generates the cubical k-chains Ck (X ) and @k : Ck (X ) ! Ck 1(X ) is a boundary operator. ld 1]) + d1] c]) @( l + l d1] @ c] @l+ l 0 0 0: Now assume that Q 2 Kn for n > 1. So b @ (@ Q) = @ (@ (I dP )) b b = @ (@ (I P )) bb b b b = @ @ I P + ( 1)dim I I @ P b b b b b = @ @ I P + ( 1)dim I @ I @ P b b b b b b b b b b = @@ I P + ( 1)dim @I @ I @ P + ( 1)dim I @ I @ P + I @@ P b b b b b b = ( 1)dim @I @ I @ P + ( 1)dim I @ I @ P : The last step uses the induction hypothesis that the proposition is true if the embedding number is less than n. then b @ (@ Q) = = = = = @ (@ l. @k g . then @ I = 0 in which case we have that each b b term in the sum is 0 and hence @@ Q = 0. On the other hand.46 The cubical chain complex for the cubical set X Rn is C (X ) := fCk (X ). l + 1].

Z0(X ) = C0 (X ) = Z.48 Let X = . Ck (X ) = 0 for all k 1. 0 1 l2 ] Example 4. Therefore. B0 = 0 and therefore.49 Let X = fx0 g Rn be a cubical set consisting of a single point. 1] 0. 1] 0] 0.2. 1.47 Let X are the elements of the subgroup Rn be a cubical set. Example 4. Since C1 = 0.. 2. 1] 1] . if k = 0 Hk (x0) = Z otherwise. : : : Example 4. 1] 1] 0. H0(X ) = Z. 0 otherwise. Since. Thus.50 Recall the cubical set = 0] 0. Ck (X ) = Z if k = 0. This allows us to immediately de ne the homology of X . Furthermore. Then x0 = l1] ln ].4. The cubical k-cycles of X Zk (X ) := ker @k Ck (X ): The cubical k-boundaries of X are the elements of the subgroup Bk (X ) := image @k+1 Ck (X ): The cubical homology groups of X are the quotient groups Hk (X ) := Zk (X )=Bk (X ): We nish this section with the computation of the homology of two extremely simple cubical spaces. De nition 4. THE ALGEBRA OF CUBICAL SETS 133 where Ck (X ) are the cubical k-chains generated by K(X ) and @k is the cubical boundary operator. Hk (X ) = 0 for all k 1. Then Ck (X ) = 0 for all k and hence Hk (X ) = 0 k = 0.

1] 0]. 1]) = 1] 0] + 1] 1] @ ( 0. 1] 0]. 1]. 1]) = 0] 0] + 0] 1] @ ( 1] 0.134 The set of elementary cubes is CHAPTER 4. 1] 0. 1] 0]. 0. 1] 1]g Thus. i. 1] 0]) = 0] 0] + 1] 0] @ ( 0. e. 1] 1]g d d b K1 ( = f 0] 0. 1]d 0]. 1] 1]) = 0] 1] + 1] 1] We can put this into the form of a matrix 2 1 0 1 03 7 6 @1 = 6 1 0 0 1 7 6 0 1 1 07 5 4 0 1 0 1 To understand Z1( 1 ) we need equation 2 1 0 6 1 0 6 6 0 1 4 0 1 This in turn means solving to know ker @1 . 1] 0]. 1]d 1]g = f 0] 0]. 0. CUBICAL HOMOLOGY K0 ( 1) = f 0] 0]. 1] 0. 1] 0]. 0]d 1]. 1] 1]g K1 ( 1) = f 0] 0. 0. 0] 1]. 1]. 0. we need to solve the 1 0 1 0 1 0 32 1 76 76 54 0 76 1 1 2 3 4 3 203 7 607 7=6 7 7 607 5 4 5 0 2 6 6 6 4 1 4 2+ 3 2+ 4 3 3 203 7 607 7=6 7 7 607 5 4 5 0 . 1] 1]g 1 ) = f 0] d0. the bases for the sets of chains b K0 ( 1) = f 0]d 0]. 1]. 0. 0. 1] 0]. @ ( 0] 0. 1] d0. 1]. 1] 1]g To compute the boundary operator we need to compute the boundary of the basis elements. 1]. 1].

H0 ( 1) = Z0( 1 )=B0( 1) is generated by 0] 0] and thus dim H0( 1 ) = 1. In particular. 0] 0] + 1] 0]. 1] 0. 1] 0. On the other hand f 0] 0] + 0] 1]. we have proven that Hk ( 1) = Z if k = 0. we need to develop more e cient methods. In particular. 0] 0] + 1] 1]g B0( 1 ): From this. THE ALGEBRA OF CUBICAL SETS 135 The only non-trivial solution to this is 1= 2= 3 = 4: Thus. Thus. Exercises . We begin with the observation that there is no solution to the equation 2 1 0 1 0 32 3 2 1 3 6 1 0 0 1 76 1 7 6 0 7 6 6 0 1 1 0 76 2 7 = 6 0 7 76 7 6 7 4 54 3 5 4 5 0 1 0 1 0 4 This implies that 0] 0] 62 B0 ( 1). While we could compute the Smith normal form we shall take a slightly di erent tack here and concentrate on equivalence classes. 1] 1]: Since. but as the reader hopefully has already seen this is a rather time consuming process. C2( 1) = 0.2. solving for the quotient space Z0 ( 1)=B0 ( 1) is a little more di cult. even if one takes a simple set such as X = 0. 1] 0. 1] 1] 0. we have that dim Z1 ( 1) = 1 and is generated by 0] 0. Furthermore. We could continue in this fashion for a long time computing homology groups. 1] 0] + 0. 1] the number of elementary cubes is quite large and the direct computation of its homology is quite tedious.4. given any element u 2 C0( 1 ) such that u 6= 0] 0] for some 2 Z one can show that u + 0] 0] 2 B0 ( 1). 1] 0. 1 0 otherwise. B1 ( 1) = 0 and hence H1( 1) = Z1 ( 1) = Z: As we learned in Chapter 3.

Run the program to nd C (X ) and H (X. 1] from the solid rectangle 0. Run the Pilarczyk programs to nd the homology of it. Zp ) for several values of p. 4. Open two gates (i. 1]3 be the boundary of the unit cube. (a) Prepare the data le for computing the chain complex C (X ) of X by the program cubchain. 3] 0. 1]. 4. (b) Determine C (T ) and compute H (P ). a possibility of passing inside from one gate to another without crossing a wall. 2) (1.6 The gure L in the le labirynth.5 Let X be a cubical set obtained by removing the center cube (1. number of pixels so it is a cubical set. Determine the cubical complex C ( 2 ) and compute H ( 2). remove two pixels) in opposite walls of the labyrinth and again run the program to nd the homology of what is left. 2) 0. (compare this set with a torus discussed in Example 4. CUBICAL HOMOLOGY 2 = bd 0.4 Let 4. Let T = bd X be its boundary.136 CHAPTER 4. Make a guess about H (X ).81. Make a guess about the solvability of the labyrinth.e. 3] 0.e. i.bmp is composed of a large but nite .

These are a very special class of topological spaces. We also know that x 2 oh (P 0) \ X and x 2 oh (Q0 ) \ X where P P 0 and Q Q0. so let x 2 CP \ CQ. Then H0(X ) is a free abelian group. H0(X ) 137 4. It says that the zero dimensional homology group measures the number of connected components of the cubical set. Since X is a cubical set there exists an elementary cube S X such that x 2 S . by Proposition 4. then CP \ CQ = . Let x.16(ii) CP is open. : : : . let CP := Q P oh (Q) \ X: Observe that P Q implies that CP = CQ. Theorem 4. Since S is convex there exists a path from P 0 to Q0 made up of edges of S . : : : . we have not said anything about the relationships between homology groups of a cubical set and topological properties of the set. P 0.3. Then there exist vertices . This implies that opS oh (P 0) \ oh (Q0) \ X . Furthermore. Also.3 H0(X ) This Chapter began with a discussion of cubical sets. Thus. For each vertex P in X . Q0 2 S . P 0 Q0 . Step 1.4. then b f Pi] 2 H0 (X ) j i = 1. and (2) using this to prove the theorem.51 Let X be a cubical set. dg forms a basis for H0 (X ). Proof: The proof consists of two steps: (1) identi ng elementary cubes with the connected components. x 2 X . Therefore. a contradiction. we need to show that CP is a connected component. The following theorem is a rst step in this direction. : : : . Finally. P 0 = Rm . y 2 CP . In particular. De ne the equivalence class P P 0 if there is a sequence of vertices R0 .. and there exist elementary edges Qk with vertices Rk 1 and Rk . However. We then moved on to the combinatorics and algebra associated with these spaces and de ned the homology of a cubical set. dg is a collection of vertices in X consisting of one vertex from each connected component of X . We will now show that if P 6 Q. if fPi j i = 1. Rm of X such that P = R0 . Let P and P 0 be vertices in X . We do this by showing that it is path connected. The proof is by contradiction.

if c = 0. b Step 2. and disjoint. Pi is a cycle for each i = 1. Observe that j@bi j CPi and therefore. Then. Therefore. . there exists an elementary cube S X such that x 2 S \ oh (P ). S is convex so there is a line segment from x to P . Rm and edges Qk as above. Observe that this implies that P 2 S . By construction. The union of the line segments and edges forms a path from x to y. To do this we need to show that d X b c= i Pi is a boundary element if and only if each i = 0. : : : d. : : : . CP is path connected. Therefore. Let P be a vertex in X . d are connected. Consider the chain m Xb c = Qk : b b Then.138 CHAPTER 4. there exists j such that P 2 CPj . Similarly. : : : . First recall that Z0(X ) = C0(X ). since i=1 @b = d X i=1 @bi b it must be that @bi = iPi. x 2 oh (P ) \ X and y 2 oh (Q) \ X . We can now conclude that the sets CPi . this implies that P Pj and hence there exist edges Qk which form a path from P to Pj . Obviously. they represent all the connected components of X . Therefore. Since P Q there exist a sequence of vertices R0 . open. However. We can write b as a sum of chains as follows d X b = bi j =1 where jbij CPi . CUBICAL HOMOLOGY P and Q such that P Q. then c 2 B0 (X ). @c = Pj P and hence k=1 b b Pj ] = P ] 2 H0 (X ): b The nal step is to show that each Pi is a distinct basis element. i = 1. there exists a path from y to Q. So assume that at least one scalar i 6= 0 and assume that c = @b for some b 2 C1 (X ). Since X is cubical.

b 0 = (@bi ) = ( i Pi) = b (Pi ) = i: . To do b this. let : C0(X ) ! Z be the group homomorphism de ned by (P ) = 1 for b b b every vertex P 2 X . Observe that b (@ Q ) = = = = b b (R1 R0) b b (R1 ) (R0 ) 1 1 0: i This implies that (@bi ) = 0 and hence b b b Thus. @ Q = R1 R0 where R0 and R1 are vertices.3. Then. H0(X ) 139 We need to show that the only way this can happen is for i = 0. Let Q be an elementary edge.4. Pi generates nontrivial homology and Pi ] 6= Pj ] if i 6= j .

1] 0. 1]. 0] 1].53 Let Q be a free face in K(X ) and let Q be a proper face of P . 0] 1].54 Let X = 0. Let Q = 0. Lemma 4.7). 1] 1]g . 1]. 1] 0.4 Elementary Collapses CHAPTER 4. then X 0 = 0] 0. Example 4. De ne X 0 := R2K0 (X ) R: Then X 0 is a cubical space obtained from K(X ) via an elementary collapse of Q through P . 1] 0. 1] 0]. 1] 1] K1 (X 0) = f 0] 0.52 Let X be a cubical set. Then. 1] R2 (see Figure 4. 1]g K1(X ) = f 0] 0. then Q P = 0. We shall now discuss a method that allows us to reduce the number of elementary cubes needed to compute the homology of the set. Proof: Assume P R. 1] 0.140 4. the elements of K1 (X ). 1] 1]g K0(X ) = f 0] 0]. 1] 0. 0. 1] 0]g K0 (X 0) = f 0] 0]. P g. De nition 4. 1] 1]g There are four free faces. 1]. Then there exists R 2 K(X ) di erent from Q and P such that Q R P . 1] 1]. even very \simple" cubical sets contain a large number of elementary cubes. Then K2(X ) = f 0. 1] 0]. Let Q 2 K(X ) be a free face and assume Q P . 0. 1]. Assume dim Q < dim P 1. 1] 0. If we let X 0 be the cubical space obtained from K(X ) via the elementary collapse of Q through P . 0. Let K0 (X ) := K(X ) n fQ. 1]. CUBICAL HOMOLOGY As the reader might have realized by now. P is not the proper face of any other cube in K(X ) and dim Q = dim P 1. Then Q R contradicting the uniqueness of P . 1] 0]. 1] 0] and 0.

In particular. 0.52. 1]. ELEMENTARY COLLAPSES 141 Observe that the free faces of K(X 0) are di erent from those of K(X 0). 1]. k 1: 0 Therefore. the domain and range of @n and @n remain the same except for n 6= k + 1. 1] 0]. through this procedure we have reduce a 2-cube to a single point. Observe that Cn(X 0) = Cn(X ) n 6= k. 1] 0]. the Hn(X 0) = Hn(X ) n 6= k + 1. Thus. dim Q0 = k 1. k 1. through P0.4. 1]. Assume dim P0 = k. Thus. 1] 0]g K0(X 00) = f 0] 0].4. This means that image @ = image @ 0 : Therefore.55 Let X 0 be obtained from X via an elementary collapse of Q0 H (X 0) = H (X ): Proof: Let @ 0 and @ denote the boundary operators on C (X 0 ) and C (X ). thus Bk (X ) = Bk (X 0). k 2. k. By Lemma 4. k 1. Zk+1(X 0) = Zk+1(X ) which implies that Hk+1(X 0) = Hk+1(X ): . k 2: b By Lemma 4. 1] 1]g On K(X 00 ) we can now perform an elementary collapse of 1] 1] through 1] 0. 0] 1] and 1] 1] are free faces with 0] 1] 0] 0. Let X 00 be the space obtained by collapsing 0] 1] through 0] 0. Then Theorem 4. 1] to obtain X 000 where K1(X 00) = f 0. 1] 0] results in the single point X 0000 = 0] 0]. respectively. P0 62 Bk (X ). 1] 0]g K0(X 00) = f 0] 0]. g A nal elementary collapse of 1] 0] through 0.52. Then. K1(X 00) = f 1] 0. k.

k i=1 where Ri 6= Q0 and ai = 1 for all i = 1.5 1 0. The reader should check that the argument is.5 0 -0. CUBICAL HOMOLOGY The cubical set X'⊂ R2 2 1.142 The cubical set X⊂ R2 2 1. Bk 2(X 0) = Bk 2(X ).5 Q P 1 0.5 2 -1 -1 0 1 2 Figure 4.5 0 -0.5 0 -0.5 0 -0.5 -1 -1 0 1 2 CHAPTER 4.5) b b X b @k P0 = Q0 + aiRi . independent of this choice. Therefore. Hk 2(X 0) = Hk 2(X ): .7: Sequence of Elementary Collapses of 0. Now 0 = @k This implies that 1 b b @k (P0) = @k 1(Q0 ) + @k b @k 1 (Q0) = @k 1 1 k X b! ai Ri : i=1 k X b! ai Ri i=1 (4. in fact.5 -1 -1 0 1 P' Q' The cubical set X'''⊂ R 2 1. It should be noted that in b b writing this equation a choice has been made for orientations of P0 and Q0.6) and hence. 1] Assume that 0. 1 R2.5 2 -1 -1 0 1 2 Q' P' 2 The cubical set X''⊂ R 2 1. : : : .5 1 0. k.5 1 0. (4.

7) b = b0 @k P0 + I X b = b0 Q0 + b0 i=1 k X i=1 b bi @k (Pi) b X b aiRi + bi @k (Pi) I i=1 b where the last equality follows from (4. Observe that this means that I X b c = biPi and hence that c 2 Zk (X 0). Then = ] for some 2 Zk 1(X ).5) and de ne 0= b0 k J X b X b aiRi + bj Sj : i=1 j =1 Then. ELEMENTARY COLLAPSES Now consider c 2 Zk (X ).5). We can write J b X b = b0 Q0 + bj Sj j =1 where Sj 6= Q0.4. This in turn implies that Zk (X 0) = Zk (X ). Q0 does not appear in either of the summations. We will do this as follows. By Lemma 4. b0 = 0. Consider 2 Hk 1(X ). by (4. Thus.4.8) .6) 3 2 k J X b X b5 0 ]=4 ai Ri + bj Sj = ] 2 Hk 1(X ): i=1 j =1 (4. Recall (4. Since i=1 Hk (X 0) = Hk (X ): The nal step is to show that there exists a group isomorphism f : Hk 1(X ) ! Hk 1(X 0). Bk (X 0) = Bk (X ).52. 0 = @k c (4. We can write I X b bi Pi: i=0 143 c= Then.

Up to this point the discussion of elementary collapses has been purely combinatorial and algebraic. This is the purpose of following discussion. bi] is an elementary interval. Then. : : : . so all that remains is to show that it is an isomorphism. 4. CHAPTER 4. : : : . So de ne f ( ]) = 0 ]: It is straightforward to check that f is a group homomorphism. Since Zk 1(X 0) Zk 1(X ) it is clear that f is surjective. xn an): Let P = TQ(Q). We have not indicated how an elementary collapse is related to a topological operation. To show it is a monomorphism assume that 1 . assume that Y H (Y ) = H (X ): From Examples 4. xn) = (x1 a1 . 2 2 Zk 1 (X ) and that f ( 1 ]) = f ( 2 ]). To simplify the formulas for the continuous maps that will be used we want to move Q to the origin.9) P = J1 Jn . x2 . The same argument that led to (4.56 Let Y can be obtained from X via a series of elementary collapses. Corollary 4. Let Q Rn be an elementary cube of the form Q = I1 In where Ii = ai . then X be cubical sets. Thus we de ne the translation TQ(x1 . x2 a2 .8) shows that 1 ] = 2 ] 2 Hk 1(X ).144 But. 1]) Z if k = 0 0 otherwise. 1] 0. (4. P has the form. CUBICAL HOMOLOGY k J X b X b aiRi + aj Sj 2 Zk 1(X 0) i=1 j =1 and thus we can view 0 ] 2 Hk 1(X 0).49 and Corollary 4.54. Furthermore.56 we can conclude that Hk ( 0.

.4. bi0 g.58 Let Q be an elementary cube. xn . Then S has the form S = J1 Jn where Ji = J ] if i = i0 . ELEMENTARY COLLAPSES 145 where Ji = 0. t) = > n 1 o tan 2 t : limt!1(x1 . di] and di 2 f0.4. Let S be the proper free face of Q such that Q0 is obtain by the elementary collapse of S through Q. 2 maxi2I nxi 1 o tan 2 t xi0 . 2 F (x1 . De ne F : P 0. 1g. then there exists i0 such that di0 = 1. xn) if t = 1. 2 maxi2I xi 2 xi0 . The R is both a free and a proper face of P . 1) P 0. and 2 fai0 . xn) < if 0 t < 1. Since Q is an elementary cube it has the form Q = I1 In where Ii = ai . Let R = K1 Kn where Ki = J1] if i = i0. then this is just restating the fact that a point is a deformation retract of an edge. Proposition 4. : : : . The case that = ai0 is left to the reader. We will present the proof in the case that = bi0 . : : : . So we can assume that dim P > 1. Then. otherwise. We leave it to the reader to check that F is continuous. : : : . i through P . 8 > (x1 . If dim P > 0. : : : .1] = id P 0 . 0) = id P . and F (P.57 Let P 0 be obtained from P via the elementary collapse of R Proof: If dim P = 1. P 0 is a deformation retract of P . Let Q0 be obtained from Q through an elementary collapse. Let I = fi j di = 1g n fi0g. 1] ! P by (4. Then Q0 is a deformation retract of Q.10) Observe that F ( . F jP 0 0. : : : . Lemma 4. bi] is an elementary interval. Proof: Let Q Rn. i otherwise.

8 Let X be the solid cubical set discussed in Exercise ??. Proof: Let X 0 be obtained by the elementary collapse of the proper free face S through the elementary cube Q. 1] 1. That this is the desired deformation retraction follows from Lemma 4. CUBICAL HOMOLOGY G(x. 1] ! X by H (x.7 Use the elementary collapses to show that the elementary cube 0. 1] ! Q by CHAPTER 4. t) if x 2 Q x otherwise. .10) and TQ is given by (4.11) where F is given by (4. 2] 0] . 2] 1] 0] . 1. 1]3 4. 2] 0] : Compute the homology of and deduce what is the homology of X . 2] 2] 0] . where G is given by (4.9). t) = G(x.59 Let X be a cubical set. Exercises is acyclic.146 De ne G : Q 0.11).57. 4. 2] 1. Then X 0 is a deformation retract of X . t) = TQ 1(F (TQ(x). Proposition 4. De ne H : X 0. We leave it to the reader to check that H is continuous. Here is an alternative way of computing the homology of X : Use the elementary collapses of X onto the simple closed curve de ned as the union of four line segments 1. t)) (4. Let X 0 be obtained from X through an elementary collapse.

Proposition 4. 1] 0]: 0] 0] 0] 0]: G1 := I1 i set 1 1 P = I1 Im 1 .4. Let De nition 4. Q1 can now be written as the union of d 1 dimensional elementary cubes.49.5 Acyclic Cubical Spaces We nish this chapter with a class of important cubical sets. Since. bi] where bi 2 f0. Let Q1 be the cubical set obtained by collapsing F through Q. 1] 0.60 A cubical set X is acyclic if Hk (X ) Z if k = 0 J := fi j Ii = 0. Proof: Let Q = I1 I2 In be an elementary cube. 1g.61 Elementary cubes are acyclic. the homology of a point 4. ACYCLIC CUBICAL SPACES 147 0 otherwise. i. 1]g: Let m = maxfi 2 J g. set G0 := I1 i and let Ii Ii 1 0] Ii+1 1] Ii+1 Im Im 0] 0] 1 0. then the result follows from Example 4. We can assume that Ii = 0. To be precise if i 2 J .) The proof is by induction on the dimension of Q. (If Q is not of this form. it is possible that d < n not all elementary intervals need be nondegenerate. then use the translation TQ to move it to the origin.5.e. Observe that F := I1 Im 1 1] 0] 0] is a free face. If dim Q = 0. Now assume that the result is true for every elementary cube of dimension less than d and that dim Q = d. those which have trivial homology.

i2 := I1 Ii1 1 Ii2 1 1 ] Ii1 +1 2 ] Ii2 +1 Im 1 0. 1g2 Once again. i2. to be precise. : : : .i2 i1. CUBICAL HOMOLOGY Q1 = P i2J Gi where 2 f0. Now observe that each Gi has a proper free face Fi1 := I1 Ii 1 1] Ii+1 Im 1 1] 0] 0] and Fi0 := I1 Ii 1 0] Ii+1 Im 1 1] 0] 0]: Let Q2 be the cubical set obtained by collapsing each Fi through Gi . 1g. each Gi1. Again. After k steps we have that Qk = P Gi1 . 1] 0] 0] Then. 1] otherwise. i2 . i2 2 J i1 < i2 2 f0. ik 2 J i1 < i2 < < ik 2 f0. Q2 = P Gi1 . 1gk where Gi1. .:::. let = ( 1. 1g2 and set Gi1 . 2) 2 f0. i2 2 J with i1 < i2.i2. Q2 can be written as a union of P and d 2 dimensional elementary cubes.148 Then CHAPTER 4.:::.ik i1 .i2 := I1 Ii1 1 Ii2 1 1 ] Ii1 +1 2 ] Ii2 +1 Im 1 1] 0] 0] which allows for an elementary collapse.i2 has a free face Fi1 .i2. for each pair i1 .ik is the elementary cube of the form J1 Jn with 8 > j ] if i = ij 2 fi1. : : : . ik g < Ji = > 0] if i 62 J : 0.

Y and X \ Y are acyclic. Since z is a cycle. As we shall see in Chapter 6 this is a simple version of a much more general and powerful theorem called the Meyer-Vietoris sequence. i . Therefore. @zX ] = n 2 Z. After. By Proposition 4.5. If X .ik has a proper free face Fi1 . 0 = @z = @ (zX + zY ) = @zX + @zY @zY = @zX : Observe that @zY . @zX ] = n 2 Z implies that .51 the assumption that X and Y are acyclic implies that X and Y are connected. We need to show that z 2 B1(X Y ).33. d iterations we have that Qd = P and by the induction step P is acyclic. b We will now show that n = 0.4. X Y is connected and hence by Theorem 4.. Proof: We will rst prove that H0(X Y ) Z.i2 . then X Y is acyclic. @zX 2 C0(Y \ X ) = Z0(Y \ X ).51. as an element of H0(Y \ X ). By Theorem 4. : : : . Thus. ACYCLIC CUBICAL SPACES 149 Furthermore. X \ Y is acyclic implies that X \ Y 6= . @z = 0. From the assumption of acyclicity. Therefore. i g > j k > 0] if i 62 Jj 1 2 < Ki = > 1] if i = m > : 0. Now consider the case of H1(X Y ). 1] otherwise. While the reduction process that we used in the previous proof is simple to implement. Proposition 4. @zX 2 C0(X \Y ) implies that @zX = P aiPi where Pi 2 K0(X \ Y ). H0(Y \ X ) Z. we would like to have conceptually easier way to conclude that a cubical set is acyclic.:::. z = zX + zY for some zX 2 C1(X ) and zY 2 C1(Y ). Gi1. The following theorem provides us with such a method.62 Assume X.ik = K1 Kn of the form 8 ] if i = i 2 fi . Therefore.i2. Let z 2 Z1(X Y ) be a cycle.51.:::. By Theorem 4. it is rather di cult to comprehend. Y Rn are cubical sets. H0(X Y ) Z.

H1(X ) = 0 which implies that there exists bX 2 C2(X ) such that @bX = b + zX .150 CHAPTER 4. @zX 2 Cn 1(Y \ X ). @z = 0. b + zX 2 Z1 (X ). n = 0. Since z is a cycle. observe that bX + bY 2 C2(X Y and @ (bX + bY ) = @bX + @bY = b + zY + b + zY = zY + zX = c: Therefore. Since @zX ] = 0 2 H0 (X Y ). Then for any Q 2 K0 (X Y ) (@ Q) = 0. Let z 2 Zn(X Y ) be a cycle. c 2 B1 (X Y ) which implies that z] = 0 2 H1(X Y ). Therefore. Finally. (@zX ) = 0. We now show that Hn(X Y ) 0 for all n > 1.2. c 2 Bn 1(X \ Y ). Since X \ Y is acyclic. It follows that zX c0 2 Zn(X ) @c = @ @zX = 0 . i. since zY 2 Cn(Y ) and zX 2 Cn(X ) we can conclude that @zY . Let c = @zX . Since c 2 Zn 1(Y \ X ). But. The same argument shows that there exits bY 2 C2 (X ) such that @bY = b + zY . Therefore. this does not imply that @zX = 0. Therefore. Hn 1(X \ Y ) 0. Then by Proposition 4. De ne the group homomorphism : C (X ) ! Z by (P ) = 1 for b i 0 b each P 2 K0(X Y ). z = zX + zY for some zX 2 Cn(X ) and zY 2 Cn(Y ). CUBICAL HOMOLOGY P a = n.e. there exists a c0 2 Cn(X \Y ) such that c = @c0 . 0 = @z = @ (zX + zY ) = @zX + @zY @zY = @zX : Of course. but X (@zX ) = ai = n: Therefore.33. there exists b 2 C1(X \ Y ) such that @b = @zX . Now observe that @ ( b + zX ) = @b + @zX = 0: Therefore. Thus. H1(X Y ) = 0. However.

which is easily checked to be acyclic. So assume that there are q elementary d dimensional cubes in X and that the proposition is true for every convex cubical set with less than q elementary d dimesional cubes and every convex set of dimension less than d. the number of intervals such that bi > ai. ai0 + 1] an. it can be written as the product of intervals. .63 If X Rn is a convex cubical set. b1 ] ai0 + 1.4. b1 ] and ai0 . Observe that for some i0 . (Note: we are not assuming that these are elementary intervals. bn]: Then. By the acyclicity of X and Y there exist c0X 2 Cn+1(X ) and c0Y 2 Cn+1(Y ) such that zX c0 = @c0X and zY c0 = @c0Y . Let X1 := a1 . and hence by induction is also acyclic. The result follows from Proposition 4. since the number of d dimesional elementary cubes in X1 and X2 are less than q. Furthermore. bn] where ai. X2 . i. Proof: Since X is a convex cubical set. bn] X1 := a1. X = a1 . then by Proposition 4. bi 2 Z. X1. bi0 ] an.e. Observe that if X is 1 dimensional. Therefore z = zX + zY = @ (c0X + c0Y ) 2 Bn(X Y ): Proposition 4. and X1 \ X2 are convex cubical sets.61 X is acyclic. If X consists of a single d dimensional elementary cube. X1 and X2 are acyclic.62. The proof will be by induction. bi0 ai0 2. then X is acyclic.5. then X is a line segment in Rn. The dimesion of X1 \ X2 is less than d. then X is an elementary cube.) Let the dimension of X be d. If not. b1 ] an . ACYCLIC CUBICAL SPACES 151 and zY + c0 2 Zn(Y ). A slightly larger collection that is topologically simple is as follows. both on the dimension of the convex set and the number of d-dimensional elementary cubes in X . Since convex cubical sets are always the products of intervals they represent a small class of cubical sets.

Proposition 4. n be a collection of starshaped sets with respect to the same point x. Proof: Since Xi is starshaped be can write Xi = Ri. Proof: Let X be a starshaped cubical set. Y is acyclic. since x 2 Ri. then X = i.j Ri.j where Ri. j . if X = iXi.j A i j \ ! = Ri. Then by the induction hypothesis.j is convex and x 2 Ri. Ti Ri. n i=1 Xi and n \ i=1 Xi are starshaped. CHAPTER 4.63 is acyclic. So assume that X = \i Xi. If k = 1 then X is convex and hence by Propostion 4. : : : . this means that X is starshaped.j and hence is starshaped. The proof is by induction on k.j is a convex and contains x. and k is the minimal number of convex sets needed to obtain X . Let Y = Sk=11 Ri. : : : k.64 A cubical set X Rn is starshaped with respect to a point x 2 Zn if X is the union of a nite number of convex cubical sets each of Proposition 4. So assume that every starshaped cubical set which can be written as the union of k 1 convex sets containing the same point is acyclic.j : i j i \ But. Again. Then.152 which contains the point x. for each j .66 Every starshaped set is acyclic. X = k=1 Ri where each Ri i is a cubical convex set. Thus. there exists x 2 X such that x 2 Ri for all i = 1. Rk is convex i S . i = 1.j .65 Let Xi. CUBICAL HOMOLOGY De nition 4. Then.j for each i. Then X = Xi 0 1 \@ = Ri.

ch (A) \ X is acyclic. by Proposition 4. region which can be written in terms of k 1 convex sets. b] T C .62. Proposition 4. because then one can nd two disjoint intervals in the family. Furthermore. because otherwise diam A 1. Proposition 4. Therefore by Proposition 4. and the conclusion follows from the previous case. Proof: First consider the case when d = 1. Exercises .4.66. Y \ Rk is a starshaped. Ri \ Rk is convex for each i = 1.63 is acyclic. Q 2 C the intersection P \ Q is non-empty. : : : . Therefore. 6= a.g: ch (A) \ X = Q2C Q: Observe that for any two elementary cubes P. Then. We cannot have b < a. Then T C is non-empty. If d > 1 then each rectangle is a Cartesian product of intervals. By the induction hypothesis it too is acyclic.67 Assume that C is a family of rectangles in Rn such that the intersection of any two of them is non-empty. X is acyclic. Then rectangles become intervals. Rn be a cubical set.68 Let X Proof: Let diam A < 1. Let a denote the supremum of the set of left endpoints of the intervals and let b denote the in mum of the set of right endpoints. Therefore . the intersection of all rectangles is the Cartesian product of the intersections of the corresponding intervals. Let A X such that Since X is cubical C := fQ 2 K(X ) jQ \A 6= .67 also T C is non-empty.5. ACYCLIC CUBICAL SPACES 153 and hence by Propostion 4. It follows that ch (A) is star-shaped and consequently acyclic by Proposition 4. k 1 and Y \ Rk = k\1 i=1 (Ri \ Rk ): Therefore.

CHAPTER 4. but X Y is 4.10 Consider the capital letter H as a 3-dimensional cubical complex. CUBICAL HOMOLOGY 4.154 not acyclic.9 Give an example where X and Y are acyclic cubical sets. . Compute its homology.

We spent most of Chapter 2 motivating the homology theory that we are using and as we did so we had to make choices of how to de ne our algebraic structures. This means that if we change our chain complex. In mathematics seeing a particular trick being employed to overcome a technicality in di erent contexts suggests that the possibility of a general procedure to take care of the problem.4. then we will have a new homology theory. The corresponding homology groups form the reduced homology of X and are denoted by ~ Hk (X ): The following theorem indicates the relationship between the two homology groups we now have at our disposal. The reduced cubical chain complex ~ ~ of X is given by fCk (X ). as the following question: Is there a di erent homology theory such that in the previous two examples we would have trivial 0th level homology? Hopefully. we showed in each case that @1 = 0. ~ Ck (X ) = Z Ck (X ) otherwise. @k gk2Z.6. given K(X ) all we need inorder to de ne homology groups is a chain complex fCk (X ). @k gk2Z where if k = 1. this question does not seem too strange. . and ~ @k := @ if k = 0. k otherwise. which means that image @1 ker : It is with this in mind that we introduce the following de nition. As was mentioned the di culty arose because H0 = Z rather than being trivial.6 Reduced Homology In the proofs of Theorem 4.69 Let X be a cubical set.51 and Proposition prop:acyclicM-V we used a speci c group homomorphism to deal with the fact that the 0-th homology group was isomorphic to Z. From a purely algebraic point of view. We can therefore. Furthermore. REDUCED HOMOLOGY 155 De nition 4. 4. The trick we employed involved the group homomorphism : C0(X ) ! Z de ned by sending each elementary cubical chain to 1.

in this case H0(X ) = 0. i=0 Now assume that X has exactly one connected component. CUBICAL HOMOLOGY ~ Hk (X ) otherwise. by Theorem 4.51 there exists d X b c0 = i Pi such that c] = In other words. dg ~ forms a basis for H0 (X ). i d c 0= i=0 i P0 .70 Let X be a cubical set. H0(X ) is a free abelian group and (~ H0(X ) Z for k = 0 Hk (X ) c0 ] 2 H0 (X ). d. Proof: Let c 2 C0 (X ). c 2 Z0(X ) implies that Pd=0 i = 0. dg is a collection of vertices in X consisting of one vertex from each connected component of X . c0 ] = 0 if and only if i = 0 for all i = 0. ~ So P assume that d 1. : : : . c 2 Z0(X ) only if (c) = 0. c 2 C0(X ). there exists b 2 C1(X ) such that c = c0 + @1 b. ~ ~ ~ Since C0 (X ) = C0(X ). (c) = (c0 + @1 b) = (c0 ) + @1 b d X b = ( i Pi = d X i=0 i=0 i: ~ Theorem 4. Thus. c 2 ~ Z0(X ) if and only if c0 = 0. : : : . if fPi j i = 0. we can write d d X b X c c0 = Pi i i P0 = i=0 d X i=0 b i (Pi i=0 c P0 ): This theorem allows us to give an alternative characterization of acyclic spaces. . Furthermore. then ~ f Pi P0] 2 H0(X ) j i = 1. But.156 CHAPTER 4. Then. Therefore. Thus. Furthermore. However. Then. : : : .

REDUCED HOMOLOGY 157 ~ H (X ) = 0: Corollary 4.71 Let X be a nonempty acyclic cubical set. then .4.6.

.7.72 The convex hull coA of a subset A of Rn is the intersection There is at least one closed convex set containing C . hence coA 6= .2 then ( 1 . n X i=0 i =1: (4. CUBICAL HOMOLOGY We present here basic de nitions and results of Simplicial Homology Theory. We shall generalize those geometric gures to an arbitrary dimension under the name simplex. given any two points x. v1 . b. A subset C of Rn is called convex if.1 Simplexes and triangulations CHAPTER 4.Keesee.12) In general. v1 . i are called barycentric coordinates of x. If this is the case. If. It is easy to see that an intersection of any family of convex sets is convex and we already know that the same is true about intersections of closed sets. the whole space Rn. Then coV is the set of those x 2 Rn which can be written as x= n X i=0 i vi . : : : . It is intuitively clear that the convex hull of two points is a line segment joining those points.73 Let V = fv0. for example a. and a convex hull of four non-coplanar points is a tetrahedron.. y 2 C . the coe cients i are not unique. d are four vertices of the unit square on Figure 2. De nition 4. the line segment x.74 A nite set V = fv0. c. y] := ftx + (1 t)y j 0 t 1g is contained in C .12 are unique. for any x 2 coV . 1 ) = 1 a + 0b + 1 c + 0d = 0a + 1 b + 0c + 1 d : 2 2 2 2 2 2 De nition 4. The proofs of the presented results and more examples may be found in most of standard textbooks in Algebraic Topology. Theorem 4. : : : . vng in Rn is geometrically independent if. Thus coA is the smallest closed convex set containing A. e.Rotman]. Munkres. of all closed and convex sets containing A. vng 2 Rn be a nite set. 0 i 1.158 4. the coe cients i in Equation 4.7 Comparison with Simplicial Homology 4. .g. a convex hull of three non-colinear points is a triangle.

Then V is geometrically independent if and only if the set of vectors fv1 v0 . : : : en+1 g where fe1 . vng 2 Rn. n-simplex spanned by vertices v0 .77 Any two n-simplexes are homeomorphic.78 Given any n 0 the standard n-simplex n is given by n := cofe1 .75 we get the following Corollary 4. The set s = coV is called simplex or. : : : . COMPARISON WITH SIMPLICIAL HOMOLOGY 159 linearly independent.75 the formula f (x) := n X i=0 i (x)wi de nes a linear continuous map f : s ! t with the contiuous inverse f 1(y) := n X i=0 i (y )vi : we will later make use of the following De nition 4. If V 0 is a subset of V of k n vertices. w1. e2 . Its special property is that the barycentric coordinates of any point x in n coincide with the cartesian coordinates x1 . Proof: . The union bd ( ) of all (k 1)-faces of a k-simplex s is called geometric boundary of s. : : : . more speci caly. the barycentric coordinates of x 2 V are continuous functions of x. It is easy to see that any linearly independet set is also geometrically independent so n is an n-simplex indeed. v1. wng be two nsimplexes. Let s = cofv0 . It is easy to verify that a point x 2 s is in bd s if and only if at least one of its barycentric coordinates is equal to zero. : : : .75 Let V = fv0 . Let i(x) be barycentric coordinates of x 2 s and i(y) barycentric coordinates of y 2 t. vn v0 g is De nition 4. v1 . e2 . vn . . vng be geometrically independent. x2 .4.7. the set coV 0 is called k-face of coV . vng and t = cofw0 . When this is the case. v1.76 Let V = fv0 . : : : xn+1 . The number n is called the dimension of V . By the de nition of geometric independence and by Theorem 4. v2 v0 . : : : . Theorem 4. v1 . From Theorem 4. : : : . : : : . : : : en+1 g is the canonical basis for Rn+1.

that this space is homeomorphic to the surface in R3 obtained by rotation of the circle (x 2)2 + z2 = 1. In this case S is called a triangulation of P .8 shows examples of subdivisions of a square to triangles. one can show. One may expect that any cubical set can be triangulated. Since S 1 S 1 2 R4. Then G is a simple closed curve Example 4. We leave the construction as an exercice. a polytope may have di erent triangulations. This set can be described as the surface of a donat.80 A subset P 2 Rn is called polytope or polyhedron if P = jSj for some simplicial complex S . it is hard to visualise it.81 By a torus we mean any space homeomorphic to the product S 1 S 1 of two circles. The intersection of any two simplexes in S is a face of each of them. The rst two are triangulations but the last one is not since the intersection of a triangle in the lower-left corner with the triangle in the upper-right corner is not an edge of the latter one but a part of it. by means of polar coordinates. Let G be the boundary of any triangle in R2. The Figure 4.79 A simplicial complex S is a nite collection of simplexes such that 1. The subset of Rn being the union of all simplexes of S is called the space of S and is denoted by jSj. 2. However . De nition 4. CUBICAL HOMOLOGY @ @ @ @ @ @ @ @ @ Figure 4. Obviously. Every face of a simplex in in S is in S . De nition 4. Neither of the above surfaces is a polytope but we shall construct one which is.160 @ CHAPTER 4. the last one is not. y = 0 about the Y -axis.8: Subdivisions of a square to triangles: the rst two are triangulations.

The bend lines divide the square to nine unitary squares.7. b] with d. 3). Thus T = G G 2 R4 is a torus. b. the reader may verify that S satis es the de nition of simplicial complex. and triangles of T obtained in this way. dg where a = (0.2 Simplicial Homology . c.9. d = (0. consider the square 0. The term simplicial complex suggests that there should be some natural structure of chain complex associated with it. COMPARISON WITH SIMPLICIAL HOMOLOGY d c 161 a b Figure 4. b = (0. In order to construct a triangulation of T we may visualise T as a square on Figure 4. 3). 3]2 = cofa.9: Triangulation of a torus hence it is homeomorphic to the unit cicle. We obtain a cylinder in R3 with a boundary of a unilateral triangle in the plane y = 0 as the base. d] with b. c]. 0). edges. c. Each of them can be divided to two triangles as shown on Figure 4. b.9 with pairs of parallel sides glued together. We bend the cylinder along the lines y = 1 and y = 2 (this cannot be done in R3 without stretching but we may add another axis) and glue the edge a. c = (3. 3).7. Let S be the collection of all vertices. Note that the four vertices a. d of the square became one. More precisely. Bend the square along the lines x = 1 and x = 2 and glue the directed edge a. c] so that the vertex a is identi ed with b and d with c. Although some vertices and edges are identi ed by gluing.4. That is not so easy to de ne 4.

In case of graphs that was easy since each edge joining vertices a. vng by the formula @k (s) = n X i=0 co(V n fvi g) : Thus. Thus C (S . : : : .2 is a special case of what we did above. The real goal however is to construct a chain complex corresponding to S with coe cients in Z as de ned in Section 3. The boundary map @k : C k (S . v1 . We put C k (S . b] or b. Z2 ) ! C k 1(S . We have the following Proposition 4. The homology of that chain comlex is the sequence of vector spaces H (S . there are many di erent ways of ordering the set of vertices. vn spanning a simplex. Z2 ) := 0 if S has no simplexes od dimension k.2. that is. As we did it with graphs. as a. @k 1@k ( ) = n XX j 6=i i=0 co(V n fvi. therefore the sum modulo 2 is equal to zero. @k gk2Z has the structure of a chain complex with coe cients in Z2 . vj g) : Each (k 1)-face of s appears in the above sum twice. Proof: For any basic element s = cofv0 . historically. a] and it was su cient to tell which vertex do we want to write as the rst and which as the last. CUBICAL HOMOLOGY due to problems with orientation which do not appear when we study cubical sets. Z2 ) = fHn(S . v1 .7. vn g. Z2 ) be the vector space generated by the set S k of k-dimensional simplexes of S as the canonical basis. : : : .82 @k 1 @k = 0 for all k. Let C k (S . this is the way homology was rst introduced. . Z2 ). We shall therefore proceed as we did with graphs in Chapter 2. the algebraic boundary of a simplex corresponds precisely to its geometric boundary. Z2 ) is de ned on any basic element s = cofv0. Z2 )g = fker @n =im @n+1 g The modulo 2 homology of graphs discussed in Section 2. In case of simplexes of dimension higher than one. we want to impose an orientation of vertices v0. : : : . in modulo 2 case. This will make de nitions much more simple and. b could be written in two ways. v1. Z2 ) := fC k (S .162 CHAPTER 4. we shall start from chain complexes with coe cients in Z2 .

vpn ) of ver- Example 4. a.10: 1. Let S be the complex consisting of s and all of its edges and vertices. An oriented simplicial complex in a simplicial complex S with one of the two equivalence clsses chosen for each simplex of S . vn] is is an equivalence class of the ordering (v0. a). (c. b]. c]. the orientation equivalence class = a. v1 . c. a] 2. v1. One may graphically distinguish the two orientations by tracing a closed path around the boundary of the triangle s following the order of vertices. The orientations of a simplex and its faces may be done arbitrarily. a. vp1 . c] 3. b. b. b. Hence we may say that the orderings which are not in the same equivalence class have the oposite orientation. a. a. a. An oriented simplex = v0 . c. a. b) and the oposite orientation 0 contains (a.4. We shall denote the pairs or oposite oriented simplexes by . c] contains the orderings (a. c. a. c].83 Two orderings (v0 . vn) and (vp0 . vi) ! (vi. the "wrong" direction of the arrows will be corrected by the minus sign in the formula for the boundary operator.7. vn) of vertices of a simplex s = cofv0. (b. : : : . c]. b). It is easy to see that for n > 0 the above equivalance relation divides the set of all orderings to two equivalence classes. b. : : : . b. Here are some among possible choices of orientations and their graphical representations on Figure 4. vng.84 Let s be a triangle in R2 spanned by vertices a. the meaning of clockwise or counterclockwise orientation is lost when we consider a triangle in a space of higher dimension. they do not need to be related. De nition 4. c. b]. a). : : : . b]. c. vi 1 ) : This de nes an equivalence relation on the set of all orderings of vertices od s. c] On the rst sight second and third orientation seem wrong since the arrows on the edges of the triangle do not close a cycle but do not worry: when we get to algebra. c]. (b. v1 . c]. However. b. (c. b. v1. b. c). a. COMPARISON WITH SIMPLICIAL HOMOLOGY 163 tices of an n-simplex s are said to have the same orientation if one can get one from another by an even number of permutations of neighboring terms (vi 1. : : : . 0. c). : : : . b. 0 or . a. b]. a. Then . The rst equivalence class gives the counter-clockwise direction and the second one the clockwise direction.

86 The group C n(S ) is a free abelian group generated by fuc^0 given by the formula 8 > 1 if = .10: Some orientations of simplexes in a triangle Let now S n be the set of all oriented n-simplexes of S . We therefore adapt the following de nition. A basis is obtained by chosing either one. : 0 otherwise. We would like to call this the group of n-chains but there is a complication: If n 0. We put C n (S ) := 0 if S contains no nsimplexes. CUBICAL HOMOLOGY c A a A A K A 6A A 2 c - A A 3 A c b a A K A 6A b a A K A 6 A A A A b Figure 4. < ~ ( ) := > 1 if = 0 . 0 . 0.85 The group of n-chains denoted by C n (S ) is the subgroup of ZSn consisting of those functions c which satisfy the identity c( ) = c( 0 ) if and 0 are oposite orientations of the same n-simplex s. Recall from Section 3.164 1 CHAPTER 4. The boundary map @k : C k (S ) ! C k 1(S ) is de ned on any basic .2 that a free abelian group ZSn generated by S n is the set of all functions c : S n ! Z.86 is related to the choice of an orientation in S . 2 S n and . Upon identi cation of the basic elements ~ with we get the identi cation of 0 with . The choice of a basis in Proposition 4. each n-simplex of S corresponds to two elements of of S n. where . 0 are oposite orientations of the same simplex. generated by basic elements ^ which can be identi ed with 2 S n. De nition 4. This set of generators is not a basis since ~0 = ~ for any pair . tions ~ = ^ Proposition 4.

H (S ) = fHn(Sg = fker @n=im @n+1 g : An important and di cult problem is to show that di erent triangulations of the same polytope have isomorphic homology comology complexes.7. That is proved by means of so called barycentric subdivisions and is too involved for the scope of this overview.7. vn] = ( 1)i v0 . vn] by the formula n X i=0 165 @k v0. numerical computations and computer graphics naturally lead to cubical sets. The concept of barycentric subdivision will appear as a by-product of the proof of Theorem 4. v1 . v1. As we already mentioned in the introduction to this chapter. Thus C (S ) := fC k (S ).7.4. v1 . COMPARISON WITH SIMPLICIAL HOMOLOGY element v0.87 in the next section. : : : . a product of a triangle by an interval is a cylinder and it has to be triangulated in order to compute the simplicial homology. that it does not depend on the choice of a representative of the equivalence class v0. @k gk2Z has the structure of a chain complex as de ned in Section 3. The homology of that chain complex is the sequence of abelian groups. : : : . 4.e. For example. : : : . : : : . A product of elementary cubes is an elementary cube but a product of simplexes is not a simplex. That feature of elementary cubes makes many proves easier and lists of data shorter. Cubical complexes have several nice properties which are not shared by simplicial complexes: 1. further subdivision to triangulations becomes arti cial and increases the complexity of data. vi 1. 2. v1 . vn] : There is a bit of work involved in showing that this formula actually de nes a boundary map: First. The reader may consult Munkers] for the proofs. one needs to show that @k 1 @k = 0. : : : . Since they already have a su cient combinatorial structure to de ne homology. vn].3 Comparison of Cubical and Simplicial Homology . one needs to show that the formula is correct i. Secondly. vi+1.

Proof: In order to keep the idea transparent we skip several technical veri cations.87 Every polytope P is homeomorphic to a cubical set. First. a linear order of real numbers imposes a choice of an orientation on each coordinate axis in Rn.166 CHAPTER 4. Each elementary 2 cube is then subdivided to 2n smaller cubes by cutting the length of each side by half. such that the restriction of h to any simplex of S is a homeomorphism of that simplex onto a cubical subset of X . there exists a homeomorphism h : P ! X . 4. given any triangulation S of P . Consider the bijection f0 of V onto the canonical basis of . a triangle is not a cubical set. canonical basis for R There is one important weak point of cubical complexes: Every polytope can be triangulated but not every polytope can be expressed as a cubical set. l]. by having written a product of inervals I1 I2 In we have implicitely chosen the ordering of the n.78. the notion of orientation in simplicial complexes is not an easy concept to learn. The notion of barycentric sudivision in the simplicial theory is much more complicated both numerically and concepionally. we have unknowingly chosen a particular orientation by having written an elementary interval as l. Secondly. We have however the following result which will help us to de ne homology of a polytope via cubical homology when we later introduce homology of a map: Theorem 4. : : : . We shall talk later about cubical subdivisions. That will be done very naturally by changing the scale on each coordinate so that the grid Zn of integer coordinates is replaced by the grid ( 1 Z)n . CUBICAL HOMOLOGY 3. As we have seen in the previous section. let S be a triangulation of P and let V = fv1. l + 1] and not l + 1. already in R. Moreover. Indeed. vN g be the set of all vertices of S . In aprticular. In other words. Step 1. The construction of h is done in two steps. We construct a homeomorphic embedding of P into a standard simplex in a space of a su ciently high dimension. Let N be the standard N -simplex in RN +1 described in De nition 4. Why does this problem not appear in the study of cubical complexes? The answer is in the fact that the de nition of a cubical set is dependent on a particular choice of coordinates in the space. v2. where X is a cubical set.

If s and t are any two simplexes of S . By the linear independence of fe1 . The injectivity of g is proved by noticing that any line parallel to L intersects N at a unique point. t) 2 RN +1. Since simplexes are closed and there are nitely many of them. x2. Moreover. s \ t is empty or is their common face so if x 2 s \ t then fs(x) = fs\t(x) = ft (x) : Thus the maps fs match on intersections of simplexes. The idea is to project a point x 2 N to a face of Y along the line L in the direction away from the origin. thus P xi = 1 and 0 xi 1 for all i. t 2 R. The image y = g(x) should have coordinates yi = xi + t for all i and some t 0. The surjectivity of g is a by-product of the construction of its inverse g 1. COMPARISON WITH SIMPLICIAL HOMOLOGY 167 X X fs( ivpi ) = i epi where i are barycentric coordinates of a point in s. it will be su cient to take X := g(P ) and de ne the homeomorphism h as the composition of f and g. Note that the supremum norm of x is jjxjj = maxfx1 . 1] such that the point x whose coordinates are givenP xi = yi t is in N . Once we do that. Consider the diagonal line L parametrised by t ! (t. eN g. Given any n-simplex s = (co)fvp0 .7. : : : . : : : xN +1 g. vp1 : : : vpn g of S . 1]N +1 which have the degenerate interval 1] on one of the components and such that any face of N is mapped ~ to a cubical face of Y . This point is in Y if 0 xi + t 1 for all i and xj + t = 1 for some j . f maps simplexes to simplexes.4. one ~ shows that P is a polytope triangulated by ff (s)g and f is a homeomorphism. We construct a homeomorphism g of N onto the cubical set Y bd 0. For this. Thus i N +1 X t = N 1 1 ( yj 1) : + j =1 j =1 j RN +1 given by f0 (vi) = ei . Step 2. t. by its construction. Let y 2 Y . e2. Recall that the barycentric coordinates of x 2 N coincide with its cartesian coordinates. we must have by 0 y t 1 for all i and N +1 (y t) = 1. In order to de ne x = g 1(y) we must nd a number t 2 0. the maps fs extend to a map ~ f : P ! P := f (P ). Thus the number t := 1 jjxjj has the desired property and the coordinates of y = g(x) are given by yi = 1 + xi jjxjj : It is clear that g is continuous. : : : . f0 extends to a map fs : s ! RN +1 by the formula . It follows that fs(s) is a n-simplex and fs is a homeomorphism of s onto fs(s). 1]N +1 consisting of those faces of 0.

: : : . If we take for granted the result of the simplicial theory saying that that the simplicial homology H (P ) of a polytope is independent on the choice of a triangulation.81 and use the homchain program to compute H (T. b]. Z2 ) for the triangulation discussed in Example 4. De ne the chain complex r C (T ). 1. t has the desired properties. 1) of Y in g is the point x := N 1 1 (1.11 De ne the chain complex C (T. d]. edges. 4. 4.14 Let K be a polytope constructed as T in Example 4. We nish this section by discussing an ineresting by-product of the above proof. A reader unfamiliar with the simplicial theory may skip it or just try to grasp the main idea. The rst homeomorphism f 1 preservers the barycentric coordinates of points in each simplex so it preserves barycenters and barycentric sudivisions. for each face Q of Y . Note that the inverse image of the vertex (1. we get H (P ) = H (X ) : In the last chapter we shall be able to arrive at this conlusion without the necessity of applying the simplicial theory. b. d] is identi ed with c.81 but with one pair of sides twisted before gluing so that the directed edge a. These observations permit to de ne a homomorphism of chain complexes C (X ) ! C (S 0 ) which induces an isomorphism H (S 0 ) = H (X ) in homology. Use the homology program to compute H (T ). c] with a.168 CHAPTER 4. Exercises 4. 4. g 1(Q) is a so called star of a vertex of N with respect to the barycentric subdivision 0N of N .13 Label vertices. : : : . 1. By continuing along these lines one can show that. Z2 ). and triangles of the triangulation of the torus in Example~efex:torus2 displayed on Figure 4. Compute H (K ). What happens if we try to use the homchain program for computing H (K.12 Prove that any cubical set can be triangulated. 1) + called barycenter of N . Z2 ) ? . CUBICAL HOMOLOGY Since 0 yi 1 for all i and yj = 1 for some j . The remaining pair of edges is glued as before.9.

COMPARISON WITH SIMPLICIAL HOMOLOGY 169 This K is called Klein bottle. Z2) ? This P is called projective plane. Compute H (P ). Note that K cannot be visualised in R3. Note that P cannot be visualised in R3.15 Let P be a polytope constructed as T in Example 4. c] with d.81 but with sides . a].7. b] and b. What happens if we try to use the homchain program for computing H (P.4. 4. d] is identi ed with c. we need an extra dimension in order to glue two circles limiting a cylinder with twisting and without cutting the side surface of the cylinder. twisted before gluing so that the directed edge a.

170 CHAPTER 4. CUBICAL HOMOLOGY .

e. then at the very least given homeomorphic spaces X and Y it should be true that H (X ) and H (Y ) are isomorphic as groups. We need to de ne a special class of group homomorphisms between the chain complexes that will induce maps on the homology groups. with a purely algebraic discussion of the latter. for the time being the set of topological spaces that we can consider is restricted to cubical sets. it must be kept in mind that F really consists of a collection of group homomorphisms Fk : Ck (X ) ! Ck (Y ): 171 . While we will use the notation F : C (X ) ! C (Y ) to represent such a map. @k g.1 Chain Maps Let X and Y be cubical sets with associated cubical chain complexes C (X ) = Y fCk (X ). 5. a chain which lies in the kernel of the boundary map. Thus. it is reasonable to expect that to de ne a map on homology one rst needs to de ne a map on the chains. As we have indicated many times by now an element of a homology group is a cycle.Chapter 5 Homology of Maps If homology is a natural invariant of a topological space. We begin. @kX g and C (Y ) = fCk (Y ). We shall do this by rst constructing a multivalued map on cubes. and then providing an algorithm for obtaining a linear map on cubical chains. To prove this we need to be able to pass from continuous maps h : X ! Y to group homomorphisms h : H (X ) ! H (Y ). i. however. Of course.

F : C (X ) ! C (Y ) is a cubical chain map if for every k 2 Z Y @k Fk = Fk 1 X @k : (5. Thus. X Y Fk c = Fk @k+1b = @k Fk b which implies that Fk (c) 2 Bk (Y ).1) Another way to describe an equality such as (5. @k g and C (Y ) = fCk (Y ).2 If F : C (X ) ! C (Y ) is a chain map. there exists b 2 Ck+1(X ) such that @k+1b = c. As was discussed in Section 5. More precisely to say that the diagram Ck (X ) ?X ?@k y Ck 1 (X ) Fk ! Ck (Y ) ?Y ?@k y Fk 1 ! Ck 1(Y ) 1 Y commutes is equivalent to saying that @k Fk = Fk X @k . then @k c = 0. Proposition 5. . @k g.172 CHAPTER 5. HOMOLOGY OF MAPS Since F is supposed to induce a map on homology. it must be the case that F maps cycles to cycles and boundaries to boundaries.1 this leads to the notion of a chain map.1 Let X and Y be cubical sets with associated cubical chain X Y complexes C (X ) = fCk (X ).1) is through the language of commutative diagrams. X Let c 2 Bk (X ). Thus which implies that Fk c 2 Zk (Y ). Then. De nition 5. then Fk : Zk (X ) ! Zk (Y ) and Fk : Bk (X ) ! Bk (Y ): X X Y 0 = @k c = Fk @k c = @k Fk c X Proof: If c 2 Zk (X ).

2 Fk ( )] 2 Hk (Y ). if ] 2 Hk (X ). By Proposition 5. F ] = Fk ] = Fk + Fk b] = Fk ( + b)] = Fk ] = F ]: We now know that cubical chain maps F.5 If there exists a chain homotopy between F and G. A collection of functions Dk : Ck (X ) ! Ck+1(Y ) is a chain homotopy between F and G if for all k Y X @k+1 Dk + Dk 1@k = G F: Restating this de nition in terms of a diagram gives % Dk Ck (X ) ?X ?@k y Ck 1(X ) G.1. = + b where b 2 Bk (X ). G : C (X ) ! C (Y ) be chain maps.4 Let F. then 2 Zk (X ).5. But. Gk Fk Ck+1(Y ) ?Y ?@k+1 y Ck (Y ) 1 ! % Dk Theorem 5. More precisely. Then.2. It is natural to ask under what conditions does F = G ? Motivate the following de ntion De nition 5. Now assume that ] = ]. G : H (X ) ! H (Y ). De ne F : H (X ) ! H (Y ) by That this map is well de ned follows essentially from Proposition 5. CHAIN MAPS F ( ]) = F ( )]: 173 De nition 5.3 Let F : C (X ) ! C (Y ) be a chain map. G : C (X ) ! C (Y ) generate homology maps F . then F = .

6 Let X R2 be the boundary of the unit square 0. Then the chain map which one can think of as being generated by rotating X by 90 degrees in a clockwise direction. Example 5. 1] 1]g : Let id : C (X ) ! C (X ) be the identity map and let F : C (X ) ! C (X ) be F0 : C0(X ) 0]d 0] 0]d 1] 1]d 1] 1]d 0] F1 : C1(X ) d 0. therefore Dk = 0 for n 1. 1] 1] 7! 1] d0. 1]. 1] d 7! 0. Then CHAPTER 5. 0] 0. By de nition it must satisfy D0@1 = F id : Let D0 : C0 (X ) ! C1(X ) .174 Proof: Let ] 2 Hk (X ). 1]. G( )] = F ( )]. 1] d 0. HOMOLOGY OF MAPS Y X G( ) F ( ) = @k+1 Dk ( ) + Dk 1@k ( ) Y = @k+1 Dk ( ) 2 Bk (Y ) Therefore. 1] d 7! 0. 1] 0] 0] d0. K1 (X ) = f 0. 1] 0] We will show that id = F by producing a chain homotopy Dk : Ck (X ) ! Ck+1(X ) from F to id . 0. 1]. 1] 0.. More precisely. 1] ! 7! 7! 7! 7! C0(X ) 0]d 1] 1]d 1] 1]d 0] 0]d 0] ! C1(X ) 7! 0] d0. 1] 1] 1] d0. This means that only D0 needs to be de ned. 1] 0]. Observe that K2(X ) = . 1] 0.

7 Assume X. CHAIN MAPS 0]d 0] 0]d 1] 1]d 1] 1]d 0] Observe that 175 7! 0] d0. Y.5. Show that p j = id Cn 1 (X ) and that j p is chain b and p(S 6 .1 Prove Proposition 5. 5. This gives an alternative shorter proof of Theorem 4. Let j : Cn 1 (X ) ! Cn(X ) be the inclusion homomorphism and p : Cn(X ) ! b Cn 1 (X ) the projection homomorphism given on generators by p(P ) := 0 b) := S if S = P . 1] = D0( 0]d 1] 0]d 0]) d = 0.1. 1] 1] 0] d0.3 Let X be a cubical set and X 0 obtained from X via an elementary collapse of a free face Q 2 Kn 1(X ) through P 2 Kn(X ) as in Theorem 4. De nition 5. 1] d 7! 0. Z are cubical sets and F : C (X ) ! C (Y ) and : C (Y ) ! C (Z ) are chain maps. Then F : C (X ) ! C (Z ) is a F) = F: The proof is left to Exercise 5. Proposition 5. Conclude from Exercise 5.8 A chain map F : C (X ) ! C (Y ) is called a chain equivalence if there exists a chain map G : C (Y ) ! C (X ) such that G F is chain homotopic to id C (X ) and F G is chain homotopic to id C (Y ) ( Exercises chain map and is an isomorphism.55. 1] = (F id )( 0] d0. 1] 0] D0@1 0] d0.1. 1]): The remaining cases are left to the reader to check. 1] 1] 7! 1] d0.55. 5.2 that H (X 0) = H (X ).7 5. 1] d 7! 0.2 If F : C (X ) ! C (Y ) is a chain equivalence then f : H (X ) ! H (Y ) homotopic to id Cn (X ) .

F (x) Y . 2. The motivation for this was given in Chapter 2. The approach we will adopt involves using multivalued maps to approximate the continuous map. x0 2Q. We did not however. A multivalued map F : X !Y ! is cubical if: 1. However. With this in mind we restrict ourselves to the following types of multivalued maps. A multivalued map F : X !Y from X to Y is ! a function from X to subsets of Y . if x. For every x 2 X .9 Let X and Y be cubical sets. sets with the same homology as that of a point. CHAPTER 5. De nition 5. then F (A) := x2A F (x): Example 5. 2) 7! 1. then F (x) = F (x0). this notion is far too general to be of use in de ning a homology theory. HOMOLOGY OF MAPS In the last section we discussed chain maps and the maps they induce on homology.e. The are a variety of possibilities. Let X and Y be cubical sets. discuss how one goes from a continuous map to a chain map.2 Cubical Multivalued maps. for every x 2 X .10 Let X = 0.176 5. De ne F : X !Y by ! 0] 7! 5] 1] 7! 0] 2] 7! 5] (0. each with it advantages and disadvantages. F (x) is an acyclic cubical set. 4] . For every Q 2 K(X ). If A X . 1) 7! 4.e.e. i. which have simple topology. Observe that since F (x) is cubical. i. get mapped to sets that have simple topology. We now want to formalize these ideas. F jQ is constant. F (x) is closed. 5]. we want to make sure that points. 2] and let Y = 5. i. In particular. In the previous chapter we introduced the notion of acyclic sets. 1] (1.

But it is obvious that there is no continuous map f : X ! Y such that f (x) 2 F (x) for all x 2 X .1.5 2 2.5. 5].5 1 1. The second is that we are interested in multivalued maps because we use them as outer approximations of continuous maps.5 0 0. continuity is de ned in terms of the pre-images of open sets. However. Recall that for single valued functions. 2] ! 5. from several points of view this is not satisfactory for our needs. the rst problem is that there are at least two reasonable ways to de ne a pre-image. However. To overcome these problems we need to introduce a notion of continuity for multivalued maps. Let F : X !Y and let B Y .1: The graph of the cubical map f : 0. The rst is that intuitively it should be clear that a map of this type cannot be thought of as being continuous.g. We want to do something similar for multivalued maps. 6 4 2 0 -2 -4 -6 -1 -0.2.5 3 Figure 5. 177 The graph of this function is given in Figure 5. CUBICAL MULTIVALUED MAPS. . The weak pre-image of B under F is ! F 1(B ) := fx 2 X j F (x) \ B 6= . Observe that F is a cubical map.

By the lower semicontinuity of F . . it is su cient to prove that F (x) F (Q). 4] R. We considered the function f (x) = (x 2)(x +1) as a map from X = 2. Let P 2 K0(X ) and let QP 2 K1(X ) be the two edges for which P 2 Q (if P = 0] or P = 2]. Then. De ne Example 5. the set F (Q) is cubical and consequently closed. Now consider x 2P . F Proposition 5. HOMOLOGY OF MAPS is open for any open U F 1(B ) := fx 2 X j F (x) B g: De nition 5. 2] R to Y = 2. Using a Taylor approximation we derived bounds on f that applied to the elementary intervals (see Table 2. which will be proved by contradiction. Proof: Since F (Q) = F (x) for x 2Q. Thus the set U := Y n F (Q) is open. V := F 1 (U ) = fz 2 X j F (z) \ U 6= . Therefore.178 while the pre-image of B is CHAPTER 5. This is equivalent to showing that x 62 V .11 A multivalued map F is upper semicontinuous if F 1(U ) is open for any open U Y and it is lower semicontinuous if the set F 1(U ) Y. F (x) = F (P ). then F (P ) F (Q). If P.7). Since F is cubical. To make this clear let us return to the discussion in Section ?? where the use of multivalued maps was rst p presented. These bounds were used to de ne F (Q) for each Q 2 K1(X ). There are two simple ways to de ne F acting on vertices. Q 2 K(X ) and P is a face of Q. age of a continuous function there is in some a canonical choice of constructing upper or lower semicontinuous cubical maps.7 and Figure 2. then set Q = Q+).12 With our goal of using multivalued maps to enclose the im- F u(P ) := F (Q+) F (QP ) P and F l (P ) := F (Q+) \ F (QP ): P u is upper semicontinuous and F l is lower semicontinuous.g is open.13 Assume F : X !Y is a cubical lower semicontinuous ! map.

Let P 2 K0(X ) and let QP 2 K1 (X ) be the two edges for which P 2 Q (if P = 0] or P = 2].5 Let X. then set Q = Q+).. de ne F (P ) := F (Q+) F (QP ) P and assume that f (x) 2 F (x) for all x 2 X .. 1]2 A : X ! Y given by R2. Exercises 5. assume that x 2 V . z 62 V . F (z) \ U = . 5. . Prove that F is upper semicontinuous. 179 So.5. it follows that x 2 Q = cl (Q). because F is cubical. (a) For P 2 K0(X ). V \ cl (Q) 6= .. Y R be cubical sets. Let f : X ! Y be a continuous function. Thus. 2]2 : A = 005 0 : 2 " # R2. (b) For P 2 K0(X ). Since x 2P and P Q. F (Q) is de ned and is an acyclic cubical set. F (z) = F (Q). Then.4 Let X = 1. CUBICAL MULTIVALUED MAPS. But V is open. Prove that F is lower semicontinuous. and hence. hence V \ Q6= . Consider the map Find a lower semicontinuous multivalued map F : X !Y with the property ! that Ax 2 F (x) for every x 2 X . a contradiction. Let Y = 2. Assume that for each Q 2 K1(X ). Let z 2 V \ Q.2. Thus. de ne F (P ) := F (Q+) \ F (QP ) P and assume that f (x) 2 F (x) for all x 2 X . (c) Show that the assumption f (x) 2 F (x) is necessary.

there exists a chain map F : C (X ) ! C (Y ) with the property for all Q 2 K(X ).1 it is su cient to indicate how a multivalued map induces a chain map. Ck (X ) = 0 . we have by Proposition 5. Of course. Q 2 K0. Theorem 5. and b b b Let Q 2 Kk (X ). choose P 2 K0 (F (Q)) and set b b F0(Q) := P: (5.2) Proof: We will construct the homomorphisms Fk : Ck (X ) ! Ck (Y ) by induction in k. we still need to indicate how this outer approximation can be used to generate homology. 2. suppose now that the homomorphisms Fi : Ci(X ) ! Ci(Y ). k 1. Then @ Q = 2 Z and Qj 2 Kk 1(X ). b jF0Qj F (Q): Furthermore. : : :. For k < 0. Q= Q and hence F (Q) = F (Q). Since.13 b jFk 1Qj j F (Qj ) F (Q) Fi 1 @i = @iFi : m P a Q for some a j bj j j =1 . Consider k = 0. 1.3) b Clearly.3 Chain Selectors. are constructed in such a way that b jFiQj F (Q) for all Q 2 Ki(K ). our purpose for introducing multivalued maps is to obtain an outer approximation for continuous functions. therefore there is no choice but to de ne Fk := 0. b jF (Q)j F (Q) (5.14 Assume F : X !Y is a lower semicontinuous cubical map. ! Then.180 5. By Section 5. For each Q 2 K0. CHAPTER 5. jF0Qj = P 2 F (Q). F 1@0 = 0 = @0 F0 : To continue the induction. HOMOLOGY OF MAPS As has been indicated since Chapter 2. Therefore. Since F is lower semicontinuous. i = 0.

there exists a chain c 2 Ck (F (Q)) such that @c = Fk 1@ Q. this procedure allows us to produce many chain maps of the type described in Theorem 5.3) of the inductive construction of F we were allowed to choose any P 2 K0 (F (Q)). where ai 2 Z.2 is called a chain selector of F .15 A chain map F : C (X ) ! C (Y ) satisfying 5. De nition 5. m. Then 6 i=1 b jF (c)j = j aiF (Qi)j i=1 m i=1 m X b jF (Qi)j . ai = 0. the homomorphism Fk satis es the property @k Fk = Fk 1@k : F (Q) = F (x) for any x 2Q. Thus 181 b jFk 1@ Qj F (Q): b induction assumption Fk 1 is a chain map.3. then Fk Q 2 Ck (F (Q)). This leads to the following de nition. so Fk 1@ Q is a cycle.14. if Q 2 Kk (X ). for any c 2 C (X ) jF (c)j F (jcj): P b Proof: Let c = m ai Qi.16 Assume F : X !Y is a lower semicontinuous cubical ! map and F is a chain selector for F . b by acyclicity. Thus. By the b Also. : : : . CHAIN SELECTORS. However.2) is well de ned. Then.5. Proposition 5. hence the set F (Q) is acyclic. hence b jFk Qj F (Q): Therefore the chain map F = fFk gk2Z : C (X ) ! C (Y ) satisfying (5. Observe that in the rst nontrivial step (5. De ne b Fk Q := c: By de nition. for all j = 1.

4) (5. they induce the same homomorphism in homology. there is no choice but to set Dk := 0. Then. b = P aiPi for some ai 6= 0 and Pi 2 Kk 1(X ). and hence. Proof: A chain homotopy D = fDk : Ck (X ) ! Ck+1(Y )gk2Z joining F to G can be constructed by induction.5) b b b c := Gk (Q) Fk (Q) Dk 1@k (Q): It followsm easily from the induction assumption that c is a cycle.13 b jDk 1@ (Q)j Consequently. and for all Q 2 Ki(X ) and c 2 Ci(Q). Thus assume k 0 and Di is de ned for i < k in such a way that @i+1 Di + Di 1 @i = Gi Fi.17 Let F. Theorem 5. m i=1 b Let Q 2 Ck (X ) be an elementary k-cube. (5. Moreover. G : C (X ) ! C (Y ) be chain selectors for the lower ! semicontinous cubical map F : X !Y .182 CHAPTER 5. For k < 0. then Pi are faces of Q b if @ Q i=1 and by Proposition 5. HOMOLOGY OF MAPS m i=1 m i=1 F (Qi) F (Qi) m i=1 = F ( Qi ) = F (jcj): The following theorem justi es the use of chain selectors. Put jDi(c)j F (Q): b jDk 1(Pi)j m i=1 F (P i) F (Q): b b b jcj jGk (Q)j jFk (Q)j jDk 1@k (Q)j F (Q): . F is chain homotopic to G.

5. Since F (Q) is acyclic. F is also a chain selector of G . Proposition 5. what procedure one uses or the amount of computation one is willing to do determines how sharp an approximation one obtains. The above theorem lets us make the following fundamental de nition. by de nition F =F =G : A fundamental property of maps is that they can be composed. Obviously. We put Dk (Q) := c0 . F is a submap of G if F (x) G (x) for every x 2 X .3. Hence. F : H (X ) ! H (Y ) is de ned by F := F : Keep in mind that the purpose of introducing multivalued maps is in order to be able to compute the homology of a continuous map by some systematic method of approximation. G : X !Y be ! lower semicontinuous cubical maps. CHAIN SELECTORS. This is denoted by F G . called the superposition of F and G by setting G F (x) := G (F (x)) for every x 2 X . and we saw this in Chapter 2. De nition 5. One easily veri es that the induction assumptions are satis ed.20 If F . Then. therefore the construction of the required homotopy is completed. G : K !L are two lower semicontinuous cubical ! maps and F is a submap of G .18 Let F : X !Y be a lower semicontinuous cubical maps.19 Let X and Y be cubical spaces and let F . In the case of multivalued maps F !X !Y and G : Y !Z we will construct the : ! ! multivalued map G F : X !Z . we conclude that there b exists a c0 2 Ck+1(F (Q)) such that @c0 = c. ! Let F : C (X ) ! C (Y ) be a chain selector of F . De nition 5. 183 It follows that c 2 Zk (F (Q)). . The homology map of F . Proof: Let F be a chain selector of F . An obvious question is how much does this matter. then F = G .

Then. De nition 5. Then by Proposition 5. We will de ne the homology of a continuous map in terms of cubical approximations. De nition 5.6) for every x 2 X .23 Let X and Y be cubical sets and let f : X ! Y be a continuous function.4 Homology of continuous maps.22 Let X and Y be cubical sets and let f : X ! Y be a continuous function. is given by f := F : . the simplest possibility would be to describe the approximation directly in terms of the image of the function and the cubes in the cubical spaces. There we started with a continuous function and used Taylor's theorem to get bounds on images of the function. A cubical approximation to f is a lower semicontinuous multivalued cubical map F : X !Y such that ! f (x) 2 F (x) (5. ! the induced homology map.16 for any Q 2 K(X ) b b j(G(F (Q))j G (jF (Q)j) G (F (Q)): Hence G F 2 C (G F ). In particular. Proof: Let F 2 C (F ) and G 2 C (G ). Recall the discussion of maps in Chapter 2. f : H (X ) ! H (Y ).21 If F : X !Y and G : Y !Z are lower semicontinuous ! ! cubical maps and G F is acyclic then (G F ) = G F . We would like our discussion of the construction of the multivalued map to be independent of the particular approximation method that is employed. HOMOLOGY OF MAPS Proposition 5. Let F : X !Y be a cubical approximation of f . These bounds were then used to construct a multivalued map.184 CHAPTER 5. But we can compose chain maps and hence (GF ) = G F = G F : 5. This leads to the following de nitions. We are nally in the position to discuss the homology of continuous maps.

f (1) = 1=3 2 0. it is easy to make a de nition. This will be the content of Section 5. 3].5. From the point of view of computations one typically wants a cubical approximation whose images are as small as possible.2.1. given cubical sets and a continuous map between them it need not be the case that there exists a cubical approximation. 5. HOMOLOGY OF CONTINUOUS MAPS. ch 0] = 0] and f (0) = 0. while ch 1] = 1].1 Cubical Approximations De nition 5. then the graph of f would no longer be contained in the graph of Mf . 1] and hence ch f (1) = 0. f (x) = x=3. In this sense Mf is minimal. Therefore. To begin with consider 0] 2 K0. 185 As is often the case. First. There are at least two questions that need to be answered before we can be content with this approach to de ning the homology of a continuous map.25 Consider the continuous function f : 0. We will deal with this problem in Section 5. 3] given by . 1]. Observe that if any cube from the graph of Mf were removed. but showing that it is well de ned or even applicable is harder. there may be many cubical approximations. Mf : X !Y . therefore Mf (0) = 0. 3] ! 0. Mf ( 1]) = 0.4.24 Let X and Y be cubical sets and let f! X ! Y be a con: tinuous function. To verify that Mf really is the minimal cubical approximation just involves checking the de nition on all the elementary cubes in 0.4. Figure 5.4. The rest of the elementary cubes can be checked in a similar manner. Example 5. The minimal approximation.4. Second. 1]. then Mf is refered to as the minimal cubical approximation. Thus.7) If Mf is a cubical map. observe that given a continuous function. On the other hand. we will need to show that all cubical approximations of a given function give rise to the same homomorphism on homology. of f is de ned by Mf (x) := ch (f (ch (x))): (5.2 indicates the graph of f and its minimal cubical approximation Mf .

.5 3 Figure 5.5 2 1. 1] K2 := 0.2: The graph of the continuous function f : 0. 4t) > > (4t 1. 1] by 8 (0. 1] ! X for t 2 0. 0) if t 2 1=4. 3] ! 0. x2 ) := (x2 ) if (x1 .5 2 2. 1) if t 2 0. 3=4] if 2 > : (4 4t. 3] and the graph of Mf . x2 ) 2 X by ( ) if (x x ) 2 K K f (x1. x2) 2 K1 K3.=1=4]=2] < t 41 (t) := > (1. Example 5.26 As is suggested by the previous de nition. 1] 3 and the map f : X ! X for (x1 .5 0 0 0. 3 4t) if t 2 1=2. x1 1 2 2 4 .5 1 0. 1] 0]: De ne the map : 0.5 1 1.186 3 CHAPTER 5.. 1] K4 := 0. 1] 1] K3 := 1] 0. Consider the cubical set X consisting of the union of the elementary cubes: K1 := 0] 0. it is not true that a minimal approximation is necessarily a cubical map. HOMOLOGY OF MAPS 2.

x2 ))) = ch (f (Ki)) = ch (X ) = X: Since X is not acyclic. Mf (x) = ch (f (ch (x))) = ch (f (Q)) F (Q) = F (x): One way to interpret this proposition is to realize that it implies that a cubical approximation for a continuous function f exists if and only if Mf is a cubical approximation. let fxng Q such that xn ! x. We have.4.26 was due to fact that the map was not acyclic. Since F is a cubical map. the answer is no. Now. the minimal approximation need not be a cubical approximation. ch (x) = Q. Mf ! is a submap of F . By continuity of f . Mf (x1 .28 Let X and Y be cubical sets and let f : X ! Y be a continuous function. then Mf is a cubical approximation. let f : X ! Y be a continuous function and let F : X !Y be a cubical approximation to f . it follows that Mf is not acyclic and consequently This example shows that given two cubical sets and a continuous map between them. x2 ) 2K i 187 not a cubical map. HOMOLOGY OF CONTINUOUS MAPS. x2) = ch (f (ch (x1 . . Proof: Let x 2 X . One can ask if there is a di erent cubical approximation for the continuous map. Proposition 5. Then f is continuous and for (x1 . Then. of course. Thus.5. The failure in Example 5. As the following proposition indicates. there exists Q 2 K(X ) such that x 2Q. given a formula for Mf .27 Let X and Y be cubical sets. f (x) 2 F (Q) which in turn implies that f (Q) F (Q). Proposition 5. therefore what remains is to understand when Mf can fail to be a cubical approximation. F (x) = F (Q) and in particular. The next proposition indicates that this is the only reason that Mf can fail to be cubical. Since x 2Q. F (x) is closed. If Mf (x) is acyclic for each x 2 X . Then.

Mf is acyclic and hence by 5. Corollary 5. Mf is de ned. y 2Q.68 is acyclic. it is easy to check . Mid X is a cubical approximation of id X and (id X ) = id H (X ) Proof: By Proposition 4. Therefore. The lower semicontinuity of Mf follows from the fact that if P is a face of Q. Conside the identity map id X : X ! X . then ! Mf is a submap of G and so de ned. b jid C (X ) (Q)j = Q = ch (Q) = Mid X (Q): Therefore. then ! Mf is a submap of F . By Proposition 5.27. F = Mf . If f has a cubical approximation. Then Mid X (x) = ch (x). id C (X ) is a chain selector for Mid X . HOMOLOGY OF MAPS Proof: Let x 2 X . let f : X ! Y be a continuous function. The fact that Mf restricted to Q is constant follows from the fact that if x. F = Mf = G : Proposition 5. if there exists a cubical approximation F : X !Y to f .28 Mf is a cubical map. Obviously f (x) 2 f (ch (x)) Mf (x). Now assume that G : X !Y is another cubical approximation to f . by Proposition 4. Finally. then f : H (X ) ! H (Y ) is well Proof: By 5. Since F is acyclic. Mid X is a cubical approximation of id X and (id X ) = Mid X : Let Q 2 K(X ). then ch (x) = ch (y). that id C (X ) induces the identity map id H (X ) on homology. then ch (P ) ch (Q). Then.29 Let X and Y be cubical sets.20. Thus.30 Let X be a cubical set.188 CHAPTER 5.16 which.

2 Rescaling So far we are able to de ne the homology map of a continuous function when a cubical approximation exists.e. Instead we take an . Mg f Mg Mf . HOMOLOGY OF CONTINUOUS MAPS. We could do the same thing here.4. 189 Proposition 5.4.e. then f : H0(X ) ! H0(X ) Proof: The homology map f : H0 (X ) ! H0 (X ) is determined by an appropriate chain map F0 : C0 (X ) ! C0 (X ). Let P 2 K0(ch (f (Q))). Q] = P ] = 1 2 H0(X ).32 Let X . We encountered this problem before in Section 2. Obviously. as was indicated in Example 5. this could be done. So let Q 2 K0(X ).21 (g f ) = (Mg f ) = (Mg Mf ) = (Mg ) (Mf ) = g f : 5. Mg and Mg f are cubical approximations. from Propositions 5. and Z be cubical sets. Which in turn can be determined by Mf acting on K0 (X ). Assume f : X ! Y and g : Y ! Z are continuous maps such that Mf .51. By de nition Mf (Q) = ch (f (Q)) which is an elementary cube. Proposition 5.20 and 5. i. is the identity map.5. and hence we have the identity map on H0(X ). If Mf is a cubical approximation of f . Unfortunately.26 not every map admits a cubical approximation. However. Y . Then. in which case f ( Q]) = P ]. that would require developing the homology theory for cubical sets de ned on fractional grids. b b b b Then. i. we could try to make the images of all elementary cubes acyclic by subdividing the domain of the map into smaller cubes. (g f ) = g f Proof: Observe that Mg f (x) = ch (g(f (ch (x)))) ch (g(ch (f ch (x)))) = Mg (Mf (x)).31 Let f : X ! X be a continous map on a connected cubical set. Therefore. but it is not necessary. we can de ne F0 (Q) = P . By Theob b rem 4. There we adopted the procedure of subdividing the intervals of our graph.

26 described a function f for which Mf was not a cubical approximation.34 Let and be a scaling vector. . 2x2 . d n ): The following properties of scalings are straighforward and left as an exercise.190 CHAPTER 5. De ne X := j . Moreover. Then. n) is another scaling vector. n ) 2 Zn and gives rise to the scaling : Rn ! Rn de ned by (x) := ( 1 x1 . if (Y ) Z . If (X ) Y .37 Let X . 4). The scaling of X by is X := X (X ) = (X ): problem involves rescaling the space X . maps cubical De nition 5. : : : . sets onto cubical sets and = . Y .35 Let X Rn be a cubical set and let 2 Zn be a scaling vector.3 shows the e ect of scaling X using the scaling vector = (4. : : : nxn ): If = ( 1 . Figure 5. De nition 5. 2 2 . : : : . and Z be cubical sets and let and be scaling vectors. Observe that if we make the domain large.33 A scaling vector is a vector of positive integers = ( 1 . then M Y X is a cubical approximation. This leads to the following notation. Proposition 5.36 Recall that Example 5. : : : . then M X is a cubical approximation. HOMOLOGY OF MAPS equivalent path based on rescaling the domain of the function to a large size. We begin by establishing that scalings are nice continuous maps in the sense that they have cubical approximations. then set := ( 1 1 . then as a fraction of the size of the domain the elementary cubes become small. 2. Example 5. The rst step in dealing with this Proposition 5. 2 .

Proof: By de nition. : : : . To show that M Y M X is acyclic. The cubical set X⊂ R 5 4 3 2 1 0 -1 -1 2 191 The cubical set X ⊂ R α 2 5 4 3 2 1 0 -1 -1 0 1 2 3 4 5 0 1 2 3 4 5 Figure 5. the map Y X is also simple. Therefore. Furthermore. However.63. Since scalings have cubical approximations they induce maps on homology. 4).5. HOMOLOGY OF CONTINUOUS MAPS. 1 1 1 X (x) := ( 1 x1 . we need to know that M X (x) = (ch ( Y (ch ( X (ch (x)))))): Therefore the acyclicity of M Y M X follows by the same argument as in Y M it induces a map . given a cubical set X and a scaling vector let X : X ! X be de ned by Obviously. The simplest way to check this is to show that their homology maps have inverses. X = ( on homology. observe that the previous paragraph. 2 x2 . for any x 2 X M X (x) = (ch ( X (ch (x))): Since ch (x) is a cube. Since Y X = X . n xn ): 1 X ) . by Corollary ?? the set M X (x) is convex and consequently acyclic by Proposition 4. it follows that X (ch (x)) is a convex cubical set.3: The space X and X where = (4. since scalings just change the size of the space one would expect that they induce isomorphisms on homology. Therefore.4.

192 CHAPTER 5.38 that M X and M X are cubical approximations.37 and Lemma 5. Thus. Now (X ).37. ( X) ( and ( X) X) =( X X X) X) = id X = id H (X ) ( X) = ( = id X = id H (X ) : As was indicated in the introduction. then ( X ) : H (X ) ! H (X ) and ( are isomorphisms. the purpose of scaling is to allow us to de ne the homology of an arbitrary continuous map between cubical sets. given a continuous map f : X ! Y and a scaling vector de ne f := f Observe that f : X ! Y . by Propositions 5. Then x 2P 2 Kk (X ). Furthermore. such that P 2 (Q). M X X and M X X are cubical approximations. X . and 5. Thus. HOMOLOGY OF MAPS X Lemma 5.38 M X : X !X is a cubical approximation of ! Proof: Let x 2 X .32. by Proposition 5. X) : H (X ) ! H (X ) ( X) 1 = ( X) : Proof: It follows from Proposition 5.30. Proposition 5. there exists M X (x) = ch ( X (ch (x))) = ch ( = Q X (P )) which is acyclic. Since X = Q 2 Km(X ). Hence. m k.39 If X is a cubical set and is a scaling vector.

: : : . As was already mentioned Figure 5. x2)) = = = = ch (f (Q)) ch ( ( 0. Then there exists a scaling vector such that Mf is a cubical approximation of f . assume for the moment that for each i = 1. 1] which is acyclic. Since dist (x.41 Let X and Y be cubical sets and f : X ! Y be continuous. : : : . if is another scaling vector such that Mf is a cubical approximation of f . ng 1= . we get from (5.32 that ( X) = ( X ) ( X) : . Also.68 that Mf is acyclic. 193 Example 5.e. n. Then = ( 1 . Consider Q = 0. X= K X . Let i := ii . HOMOLOGY OF CONTINUOUS MAPS.3 shows X and f : X ! X . Moreover. i. Similar checks at all the points on X shows that Mf is acyclic and hence Mf is a cubical approximation.5. y 2 K ) dist (f (x). Therefore it follows from Proposition 5. Consider = (4. Now consider Mf . i j i . 1]) 0] 0. Now assume that the scaling vector is such that Mf is also a cubical approximation. 1] 4]. : : : . y) diam ch (x) 1. n) is a scaling vector.40 To indicate the relationship between f and f we return to Example 5. Let (x1 . Then Mf (x1 . Mf is a cubical approximation of f . Proposition 5. Clearly. x2) 2Q. f (y)) 1 (5. 1=4])) ch ( 0] 0.4. then f ( X) = f ( X) Proof: Choose > 0 such that for x. 4).8) 2 and let be a scaling vector such that minf i j i = 1.8) that diam f (ch (x)) 1: 2 Therefore it follows from Proposition 4. x2 ) = ch (f (ch (x1 .26.

Let be a scaling vector such that Mf is a cubical approximation to f . n. We can now give the general de nition for the homology map of a continuous function. HOMOLOGY OF MAPS On the other hand we also have f =f X . if it is not true that i j i for each i = 1. Lemma 5. The nal issue we need to deal with involves the composition of continuous functions. then let = . Then.41.42 Let X and Y be cubical sets and let f : X ! Y be a continuous function. hence Mf (M X (x)) = ch (f (ch ( X (ch (x))))) = ch (f (ch (x))) = Mf (x): Therefore we get from Proposition 5. we need to reconcile this de nition of the homology map with that of De nition 5. If Mf and M M Y Mf is a cubical map. Then f = f and hence the two de nitions of f agree. So assume that Mf is a cubical approximation of f . De nition 5. : : : .32 that f =f ( X ) and consequently f ( X) = f ( X ) ( X) = f ( X) : Finally.23. We will need the following technical lemma. Let be the scaling vector where each i = 1. However. then . f : H (X ) ! H (Y ) is de ned by f =f X: By Proposition 5. By what we have just proven f ( X) = f ( X) = f ( X) which settles the general case.43 Let X and Y be cubical sets and let f : X ! Y be continuous. this de nition is independent of the particular scaling vector used. Y Let be a scaling vector.194 CHAPTER 5. f are cubical approximations.

68 Mg M f (x) = ch g ch f ch (x) is acyclic. i. y in X dist (x. select a scaling vector such that Mf and Mh are cubical approximations.68 that h = (g K f ) = g ( K f ) : . g (y)) 1 : 2 Similarly. Observe that M Y Mf (x) = ch ( Y (Mf (x))) = Y (Mf (x)): Since Mf (x) is acyclic.9) dist (x.5. for any x 2 K diam (g ch f ch (x)) < 1: Therefore by Proposition 4.39 that Y (Mf (x)) is Proposition 5. y) 2 ) dist (g (x). Then (g f ) = g f Proof: Select a scaling vector such that Mg is a cubical approximation and for any x.4.44 Assume f : X ! Y and g : Y ! Z are continuous maps between cubical sets.e. y) 2 ) dist ( f (x). and for any x.9 and 5. by 5. HOMOLOGY OF CONTINUOUS MAPS. it follows from Proposition 5. It follows from Proposition 4. Proof: We only need to verify that M Y 195 also acyclic.10. Hence (g f ) = g f : By Lemma 5.10) 2 Then the maps f and g ( f ) = h have cubical approximations. y 2 Y (5.43 we also have that ( K f ) = K f : Let h := g f . Mf is acyclic. f (y)) 1 : (5. Moreover. the composition Mg M f is acyclic.

x2) 2 K1 > 2 > (x ) < 1 if (x1 . b) Find a chain selector of Mf . You may either compute it by hand or use the homology program for that. we shall prove the following theorem. 5. In this section we shall pursue the question of when do two continuous maps induce the same homomorphism on homology. 5.3 was a partial answer in that we showed that H0 (X ) counts the number of connected components of X .34 5.6 Prove Proposition 5. (g f ) = = = = = Exercises CHAPTER 5. Section 4. . x ) 2 K 2 1 2 > : (1 x1 ) if (x1 .26. x2) := > (1 x ) if (x . and f be as in Example 5. HOMOLOGY OF MAPS h h ( K) g ( K f ) ( K) g ( K ) f ( K) g f: a) Verify that the scaling by := (2. 2) is su cient for Mf to be a cubical approximation of f . In particular.7 for the map given by 8 (x ) if (x1 . What is missing is how these algebraic objects relate back to topology.8 Do the same as in Exercise 5. .3 Homotopy Invariance of Maps We now have a homology theory at our disposal. Given a cubical set X we can compute its homology groups H (X ) and given a continuous map f between cubical sets we can compute the induced map on homology f . c) Compute the homology map of f .7 Let X . x2) 2 K3 4 5.4. x2) 2 K2 f (x1 .196 Hence.

What makes this simple is that an approximation for the homotopy provides an approximation for both f and g. HOMOLOGY OF CONTINUOUS MAPS. 1]. However.45 Let X and Y be cubical sets and let f. De ne H0 : X !Y by ! ! H0 (Q) := H(Q 0.4. 1) = g(x).46 H0 : X !Y is a cubical approximation to both f and g. Observe that is a homotopy between f and . 0) for all x 2 X . by Proposition 5. 1]!Y to . for any x 2 X . f (Q 0]) H(Q) and in particular. we can obtain a cubical approximation for an appropriate scaling of .5. By de nition f g implies that there exists a continuous function : X 0. topic maps. Observe that X 0. Therefore. f (x) 2 H0 (x). f (x) 2 H(x. 0) = f (x) and (x. H(Q) H(Q 0. Then 197 Theorem 5. 1]).13. The rst is trivial. ! Proof: Clearly. However. A similar argument holds for g.47 If the homotopy from f to g has a cubical approximation then f =g : f = H0 = g : Proof: By de nition This was the easy case. for every Q 2 K(X ). Corollary 5. Assume for the moment that there exists a cubical approximation H : X 0. choose a scaling vector such that M is a cubical approximation. need not admit a cubical approximation. 1]) Lemma 5. g : X ! Y be homof =g : We shall break up the proof into two cases. Therefore. Q is a face of Q 0. With this in mind. but contains the essential observation. The second case is merely an elaboration of the rst needed to overcome some technical di culties. 1] is a cubical set. Of course. as was made clear in the previous section. 1] ! Y such that (x.

198

CHAPTER 5. HOMOLOGY OF MAPS

g . If X Rn then 2 Zn+1 . Let = ( 1; : : : ; n) and let n+1 = k. Set Z = (X ). Then (X ) = Z 0; k]: For i = 0; : : : ; k, let fi : X fig be de ned by fi (x) = f (x; i) and let i : X i; i+1] ! Y be given by i (x; s) = (x; s). Then, i is a homotopy from fi to fi+1. However, by assumption i has a cubical approximation. Therefore, fi = fi+1 and hence, f0 = fk . Now observe that f = f0 and g = fk , therefore f =g : This proves Theorem 5.45 in the general case.

Exercises

5.9 In one of the two previous exercises you should get the trivial homology map. Prove this in a di erent way, by showing that your f is homotopic to a constant map.

5.5. LEFSCHETZ FIXED POINT THEOREM

199

**5.5 Lefschetz Fixed Point Theorem
**

We are now in the position to prove one of the most important results in algebraic topology, the Lefschetz xed point theorem. Let f : X ! X be a continuous map. x 2 X is a xed point of f if f (x) = x. The Lefschetz theorem gives conditions on f that imply that f has a xed point. We need a few algebraic preliminaries before we can state and prove the theorem. Let A = aij ] be an n n matrix. The trace of A is de ned to be the sum of the diagonal entries, i.e. n X tr A = aii : It is easy to check that if A and B are n n matrices, then tr AB =

i=1

X

i;j

aij bji = tr BA:

Let G be a nitely generated free abelian group and let : G ! G be a group homomorphism. Since G is free abelian, it has a basis and for a particular choice of basis can written as a matrix A. So in this case de ne tr = tr A: To check that this is a well de ned concept, let fb; : : : ; bn g be a di erent basis for G. Let B : G ! G be the isomorphism corresponding to the change of basis. In this second basis the matrix representation of is given by B 1 AB . Thus, tr (B 1AB ) = tr (B 1(AB )) = tr ((AB )B 1 ) = tr A: We will need to apply these ideas in the context of homology groups. Consider a free chain complex fCk (X ); @k g and a chain map F : C (X ) ! C (X ). Let Hk (X ) be the induced homology groups and f : H (X ) ! H (X ) the induced homology map. Since Ck (X ) is a free abelian group, tr Fk is well de ned for each k. However, the homology groups Hk (X ), while abelian need not be free. Let Tk (X ) denote the torsion subgroup of Hk (X ). Then, Hk (X )=Tk (X ) is free abelian. Furthermore, f : H (X ) ! H (X ) induces a homomorphism

k

: Hk (X )=Tk (X ) ! Hk (X )=Tk (X ):

k

Thus,

k

can be represented as a matrix, and hence tr

is well de ned.

200

**CHAPTER 5. HOMOLOGY OF MAPS X
**

k

**De nition 5.48 Let X be a cubical set and let f : X ! X be a continuous
**

map. The Lefschetz number of f is

L(f ) :=

( 1)k tr k :

Theorem 5.49 Lefschetz Fixed Point Theorem Let X be a cubical set and let f : X ! X be a continuous map. If L(f ) 6= 0, then f has a xed point.

This theorem is an amazing example of how closely the algebra is tied to the topology. To prove it we need to understand how to relate the topology in the form of the map on the chain complexes to the algebra in the form of the induced homology maps on the free part of the homology groups. We begin with a technical lemma.

Lemma 5.50 Let G be a free abelian group, let H be a subgroup and assume that G=H is free abelian. Let : G ! G be a group homomorphism such that (H ) H . Then, induces a map 0 : G=H ! G=H and tr = tr 0 + tr jH :

Proof: The rst step is to understand 0 . Since G is free abelian, and H is a subgroup, H is also free abelian. Let f 1; : : : ; k g be a basis for H and let f 1 + H; : : : ; n + H g be a basis for G=H . Then, 0 is de ned by

0(

i + H) =

( i):

It is left to the reader to check that 0 is a well de ned group homomorphism. Given our choice of basis we can represent 0 as a matrix B = bij ]. In particular, n 0 ( j + H ) = X bij ( i + H ): Similarly, jH : H ! H has the form

i=1 k X i=1

jH ( i) =

and so we can write jH = A = aij ].

aij

i

**5.5. LEFSCHETZ FIXED POINT THEOREM
**

Since G = G=H H , f 1 ; : : : ; k ; 1; : : : ; ng is a basis for G. Thus, ( j) = ( j) =

k X i=1 n X i=1

201

aij

i

bij i = h

where h 2 H . This means that as matrix has the form

# A = 0 B :

"

Clearly, tr = tr 0 + tr jH .

Theorem 5.51 (Hopf trace theorem) Let fCk (X ); @k g be a free chain complex and F : C (X ) ! C (X ) a chain map. Let Hk (X ) denote the corresponding homology groups with torsion subgroups Tk (X ). Let k : Hk (X )=Tk (X ) !

Hk (X )=Tk (X ) be the induced homomorphism. Then X k X ( 1) tr Fk = ( 1)k tr k :

k k

Proof: We will use the notation from Section 3.7 where Wk (X ) denotes the weak boundaries. Recall that

Bk (X ) Wk (X ) Zk (X ) Ck (X ): Furthermore, since F is a chain map, each of these subgroups is invariant under Fk , i.e. Fk (Bk (X )) Bk (X ), Fk (Wk (X )) Wk (X ), etc. From Lemma 5.50 Fk induces maps Fk jWk (X ) : Wk (X ) ! Wk (X ); Fk0 : Zk (X )=Wk (X ) ! Zk (X )=Wk (X ) Fk00 : Ck (X )=Zk (X ) ! Ck (X )=Zk(X ): From Lemma 3.63 and the following comments we have that for each k, Zk (X )=Wk (X ) and Ck (X )=Zk(X ) are free abelian groups. Therefore, applying Lemma 5.50 twice gives tr Fk = tr Fk00 + tr Fk0 + tr Fk jWk(X ) : (5.11)

the diagonal entries of F indicate how the duals of elementary cubes are mapped to themselves. However.63 Zk (X )=Wk (X ) = Hk (X )=Tk (X ) and furthermore under this isomorphism.15) and hence mj aij = bij mi and in particular miaii = biimi . HOMOLOGY OF MAPS However. l g for Wk (X ) and integers m1.12) can be written as tr Fk = tr Fk 1 jBk 1(X ) +tr k + tr Fk jWk (X ) : (5. Applying this to (5. The Hopf trace formula provides us with a means of relating a chain map F : C (X ) ! C (X ) for f with L(f ). : : : . before beginning the proof let us discuss the basic argument that will be used. (5. This is what we will prove. The Hopf trace formula is the key step in the proof of the Lefschetz xed point theorem. Therefore. Observe that l X Fk jWk (X ) ( j ) = aij i (5. So multiplying (5. However. If f has no xed points then the image of a small cube will not intersect itself and so the . Observe that an equivalent statement to the Lefschetz xed point theorem is the following: if f has no xed points. from Lemma 3. ml . such that fm1 1.13) We will now show that tr Fk jWk (X ) = tr Fk jBk (X ) . tr Fk jWk (X ) = tr Fk jBk (X ) .14) and i=1 Fk jBk (X ) (mj j ) = l X i=1 bij mi i (5. In particular. ml l g is a basis for Bk (X ). (5. if we could show that tr F = 0.16) by ( 1)k and summing. Both these maps are just restrictions of Fk to the appropriate subspaces. An therefore.12) Similarly.202 CHAPTER 5. Of course.13) give tr Fk = tr Fk 1 jBk 1 (X ) +tr k + tr Fk jBk (X ) : (5.11) becomes tr Fk = tr Fk00 + tr k + tr Fk jWk (X ) : (5. Therefore. Ck (X )=Zk (X ) is isomorphic to Bk 1(X ) and under this isomorphism Fk00 becomes Fk 1 jBk 1 (X ) . : : : .16) The proof is nished by multiplying (5. As was done explicitly in Section 3. Fk becomes k . : : : . then L(f ) = 0.15) for appropriate constants aij and bij . then it would be clear that L(f ) = 0. the easiest way to check that tr F = 0 is for all the diagonal entries of F to be zero.7 there exists a basis f 1.14) by mj must give rise to (5.

> 0.5.: Let y 2 ch (x). X (y ) Therefore. This implies that jjy xjj 1 and hence jj X (y) X (x)jj < : f X (x)jj =3: By the de nition of followed by the triangle inequality we have < jj X (x) f X (x)jj < jj X (x) f X (y)jj + jjf X (y) f X (x)jj This implies that jj X (x) f X (y)jj > 2 =3 and therefore. is presented in the reverse order. jjf jj X X (x) X f X (y )jj = jjx X f X (y )jj > 12 3 = 4: . Let := min jjx f (x)jj: x2X Since we are assuming that f has no xed points and X is cubical.5. LEFSCHETZ FIXED POINT THEOREM 203 diagonal entries are zero. Mg (x) \ ch (x) = . Assume f has no xed points. Let be a scaling vector with the property that i > 1 for each i. We want to show that L(f ) = 0. consider g := X f X : X ! X : We will now show that for any x 2 X . Proof of Lefschetz Fixed Point Theorem. which as is often the case in mathematics. Similarly. With these constants in mind. since X is cubical there exists > 0 such that jjx yjj < ) jjf (x) f (y)jj < =3: Set := minf . =6g. With this argument in mind we turn to the proof. The rst step is to establish some constants that will used in the proof.

Finally. Let f : X ! X be continuous. but b b b jG(Q)j \ Q = .. Therefore. f : H0(X ) ! H0 (X ) is the identity map. Then.204 CHAPTER 5.41 we can assume that was chosen large enough that Mg is a cubical approximation of g. Therefore. But. tr G = 0 and therefore by the Hopf trace formula. . HOMOLOGY OF MAPS This inequality holds for all y 2 ch (x). In particular. and therefore Mg (x) \ ch (x) = . f has a xed point. by Proposition 5. by Proposition 5. and therefore the diagonal entries of G are zero. Let G : C (X ) ! C (X ) be b a corresponding chain map. L(g ) = 0.52 Let X be an acyclic cubical set. by Proposition 5. Proof: Since X is acyclic. The standard basis for C (X ) is K(X ). L(f ) = 1. Observe that the this argument is true for any su ciently large. the only nonzero homology group is H0 (X ) = Z.39 L(f ) = L(g ): Theorem 5. for all Q 2 K(X ).31.

: Yellow shows the set E := ch ( 4.Chapter 6 Homological Algebra We nished the previous chapter with the Lefschetz xed point theorem that allowed us to prove the existence of xed points of maps from the homology map. Unfortunately. 4] (2. 4] 4. but as the following example indicates it has its limitations. In Figure 6. We leave it to the reader to check that it is impossible to nd a cubical set X that contains a neighborhood of the origin such that f (X ) X . So lets study the problem is little further. f (X ) 6 X . consider X := fx 2 R2 j jjxjj 4g. However. 4]) \ X = . 4] (2. 2) 4. But to talk about a xed point we need to have a map of the form f : X ! X . To begin with R2 consists of an in nite number of cubes and hence. so we could. there is no direct way to apply the Lefschetz theorem to detect this xed point. Obviously. Let f : R2 ! R2 be the linear map " # 2 0 : R2 ! R2 f = 0 1=2 Obviously the origin is a xed point. there is a problem in that f ( 4. As we noted before this is a remarkable theorem.1 the set X := fx 2 R2 j jjxjj 4g is indicated in red and its image under f in blue. 2) 4. Unfortunately. for example. Consider for the moment the following almost trivial example. 4] 4. We know that the origin is the xed point. the Lefschetz xed point theorem is too nice a tool to give up trying to extend it to an example such as this. is not a cubical set. We could try to get around this problem by restricting the domain and range of the function. 4]): 205 .

This leads to the notion of relative homology. Since K(X ) b b is a basis for C (X ) and K(E ) K(X ).8] × [-2. E ) := Ck (X )=Ck (E ): . Thus we can make the following de nition.4] × [-4. De nition 6.4] 10 5 0 -5 -10 -10 -5 0 5 10 Figure 6.1 Let X and E X be cubical sets.1 Relative Homology Let X and E X be cubical sets. 6. the xed point that is of interest lies in X n E . the quotient group C (X )=C (E ) is a free abelian group. They generate the sets of elementary cubes b K(X ) and K(E ) which in turn de ne the chains C (X ) and C (E ). This suggests that we try to develop a homology theory that begins with the set X but \ignores" the set E .-2] × [2.4] 10 10 The set f(X)=[-8.1: The image of the linear map f . The relative chains of X modulo E are the free abelian groups Ck (X.2] 5 5 0 0 -5 -5 -10 -10 -5 0 5 10 -10 -10 -5 0 5 10 The set E=[-4.206 CHAPTER 6. HOMOLOGICAL ALGEBRA Of course. The set X=[-4.

X connected implies that for any pair P. H0(X.1) b b 0 = Q] = P ] 2 H0(X. Since @0 = 0. E ) we begin by examining the associated set of cycles Z0(X. E ). E ). E ) = 0. E ) ! Ck 1(X. E ) := ker @k : Ck (X. E ). Q 2 C0(X ). E ) for any P 2 K0(X ). E ) = 0: Proof: To compute H0 (X. E ) := image @k+1 : Ck+1(X.1) b E 6= . the relative k-boundaries.. by (6. hence there exists Q 2 K0(E ). E )=Bk (X. 0 = Q 2 Z0(X. . E ) ! Ck (X. E ) is the boundary map induced by the standard boundary map on Ck (X ). there exists c 2 C1(X ) such that b b @c = P Q: (6. E ).1. Therefore.2 Let X be a connected cubical set and let E be a non-empty H0(X. @k g where @k : Ck (X. E ). RELATIVE HOMOLOGY The relative chain complex of X modulo E is given by 207 fCk (X. Therefore. E ) ! Ck 1(X. E ): cubical subset of X . E ) := C0(X )=C0(E ): b b From the proof of Theorem ??. Proposition 6. Zk (X. The relative chain complex gives rise to the relative k-cycles. Z0(X. E ) := Zk (X. By de nition. E ) = C0(X.6. and nally the relative homology groups Hk (X. Bk (X. Then.

208

CHAPTER 6. HOMOLOGICAL ALGEBRA

Example 6.3 Let X = 1; 1] R and let E = f 1g. HK (X; E ) = 0 for all k 2, since Kk (X ) = ;. By Proposition 6.2, H0 (X; E ) = 0.

Therefore, all that remains to be computed is H1(X; E ). Observe that C1(X ) Z2 with a basis given by f d0]; 0; 1]g. Since C1(E ) = 0, C1(X; E ) = 1; d C1(X ). The computation of C0(X; E ) is a little more interesting. The standard basis for C0(X ) is f d ; c g, while the corresponding basis for C0 (E ) 1] 0] d g. Therefore, is f 1] C0(X; E ) = Z and generated by c Using these bases the matrix representation for @1 : 0]. C1(X; E ) ! C0(X; E ) is @1 = 1 1]: d The chain d0]+ 0; 1] 2 C1(X; E ) clearly generates the kernel of @1 . There1; fore, H1(X; E ) Z: As will become clear as we progress, relative homology groups are a very powerful tool. So much so that we want a simple shorthand notation for discussing pairs of cubical sets. With this in mind the statement that (X; E ) is a cubical pair or a pair of cubical sets means that X and E are cubical sets and E X .

Example 6.4 Let X = 3; 3] and E = 3; 1] 1; 3]. The exact same arguments as in the previous example show that Hk (X; E ) = 0 if k = 1. 6 d 6 with a basis given by f i; i + 1] j Let us compute H1(X; E ). C1(X ) = Z i = 3; : : : ; 2g. In contrast to Example 6.3, C1(E ) = Z4 with a basis f i; id 1] j i = 3; 2; 1; 2g. This implies that a basis for C1(X; E ) con+ sists of the equivalence classes containing f d0]; 0; 1]g. Repeating this 1; d type of argument on the level of the 0-chains we see that C0(X; E ) = Z with

a basis consisting of the equivalence class de ned by c Observe that on the 0]. level of relative chains we have the same chain complex as in Example 6.3. Therefore,

H ( 3; 3]; 3; 1] 1; 3]) = H ( 1; 1]; f 1g):

One can ask whether these two examples are merely a coincidence or represent a deeper fact. Since in the relative chains of the pair (X; E ) one quotients out by those elementary chains which lie in the subspace, it seems

6.1. RELATIVE HOMOLOGY

209

reasonable to conjecture that if one adds the same cubes to both X and E , then the group of relative chains does not chain and hence the homology should not change. Theorem 6.5, presented shortly con rms this, though at rst glance its statement may appear somewhat di erent. Of course to compare the relative homology groups of di erent pairs we need to be able to talk about maps. So let (X; E ) and (Y; B ) be cubical pairs. Let f : X ! Y be a continuous map. The most basic question is whether f induces a map from H (X; E ) to H (Y; B ). If this is to be the case then there must be an associated chain map F : C (X; E ) ! C (Y; B ). However, this can only occur if F (C (E )) C (B ). This leads to the following condition.

f : (X; E ) ! (Y; B )

is a continuous map between cubical pairs if f : X ! Y is continuous and f (E ) B . To generate a map on the level of relative homology, i.e. f : H (X; E ) ! H (Y; B ) we proceed as before. f : X ! Y is continuous and so for an appropriate scaling vector , Mf : X !Y is a cubical approximation. Since ! f (E ) B and B is cubical, Mf (E ) B . Now let F : C (X ) ! C (Y ) b be a chain selector for Mf . For any Q 2 K(E ), jF (Q)j Mf (Q) B , and hence F (C (E )) C (B ). Thus, F induces a chain map between the relative chain complexes, i.e. with a slight abuse of notation we can write F : C (X; E ) ! C (Y; B ). Then we de ne f : H (X; E ) ! H (Y; B ) by

f ( ]) := F ( )]:

Theorem 6.5 (Excision Isomorphism Theorem) Let (X; E ) be a cubical set. Let U E be a representable set such that E n U is a cubical set. Then, the inclusion map : (X n U; E n U ) ! (X; E ) induces an isomorphism e : H (X n U; E n U ) ! H (X; E ): Proof: Since : (X n U; E n U ) ! (X; E ) is the inclusion map, M (Q) = Q for every Q 2 K(X n U ). Thus, the inclusion map I : C (X n U ) ! C (X ) is a chain selector for Me. Let : C (X ) ! C (X; E ) be the projection map. Then I : C (X n U ) ! C (X; E ) is surjective. To see this observe that a

basis for C (X; E ) consists of all

b b b b Q 2 K(X ) n K(E ) K(X n U ):

210

CHAPTER 6. HOMOLOGICAL ALGEBRA

b Furthermore, the kernel of I is exactly K(E n U ). Therefore, I induces an isomorphism e : C (X n U )=C (E n U ) ! C (X )=C (E ) and hence e : H (X n U; E n U ) ! H (X; E ) is an isomorphism. We began this chapter with a simple example of a linear map on the plane and asked the question whether it is possible to detect the xed point using algebraic topological methods. Referring to Figure 6.1, we see that X = 4; 4] 4; 4] is the region we want to study. Unfortunately, f (X ) 6 X . However, we identi ed E = 4; 2] 4; 4] 2; 4] 4; 4] as the smallest cubical set with the property that if x 2 X and f (x) 62 X , then x 2 E . Thus, E is a cubical representation of the exit set for X , i.e. those points which leave X under one iteration. As was noted before, f (X ) = 8; 8] 2; 2]. Clearly, f (E ) = 8; 4] 2; 2] 4; 8] 2; 2]. Combining these two observations, we can write f (X ) X f (E ): So let Y = X f (E ) and let B = E f (E ). Then f : (X; E ) ! (Y; B ) is a continuous map between cubical pairs. Now let U = Y n X . This is a representable set and B n U = E which is a cubical set. Therefore by the Excision Isomorphism Theorem e : H (Y n U; B n U ) ! H (Y; B ) is an isomorphism. But (Y n U; B n U ) = (X; E ), therefore, e 1 : H (Y; B ) ! H (X; E ) is an isomorphism. De ne f(X;E) : H (X; E ) ! H (X; E ) by f(X;E) := e 1 f . We now have a map, at least on the level of homology, that goes from a space to itself and we can hope to develop a Lefschetz xed point theorem for this map that would tell us about the existence of xed points for f restricted to X n E .

Exercises

6.2. EXACT SEQUENCES

211

**6.1 Let Q 2 Kq be an elementary cube. Let E = fP 2 Kq 1 j P is a proper face of Qg. Prove that Hk (Q; E ) = Z if k = q
**

0 otherwise.

6.2 Let f : (X; E ) ! (Y; B ) be a continuous map between pairs. Choose a scaling vector such that Mf : X !Y is a cubical approximation. Prove ! that Mf (E ) B . 6.3 Let X = 4; 4] 4; 4], E = 4; 2] 4; 4] 2; 4] 4; 4] and

f = 2 102 : R2 ! R2: 0 =

Compute H (X; E ) and f(X;E) .

"

#

6.2 Exact Sequences

We nished the last section with a suggestion that we were close to being able to develop a Lefschetz xed point theorem for pairs of spaces. However, if the reader solved Exercise 6.3, then it is clear that our ability to compute relative homology groups, is rather limited. Thus, before continuing our quest for a xed point theorem we will look for more e cient methods of computing relative homology groups. Given a pair of cubical sets (X; E ), ideally, we would have a theorem that by which we could determine H (X; E ) in terms of H (X ) and H (E ). As we shall see in Section 6.3 such a theorem exists, but before we can state it we need to develop some more tools in homological algebra. From the algebraic point of view, homology begins with a chain complex fCk ; @k g which can be thought of as an sequence of abelian groups and maps

+1 k : : : ! Ck+1 @k! Ck @! Ck 1 ! : : :

with the property that

image @k+1 ker @k : A very special case of this is the following.

212 CHAPTER 6. image g1 = ker = G0.9 Assume that 3 2 1 G3 g! G2 g! G1 g! G0 is an exact sequence. ker = G0 . If the sequence has a rst or last element.7 G1 g! G0 ! 0 is an exact sequence if and only if g1 is an is exact at G2 if epimorphism. 1 Lemma 6.e. i. 2. HOMOLOGICAL ALGEBRA De nition 6. 1 Proof: ()) Assume that G1 g! G0 ! 0 is an exact sequence. then it is automatically exact at that group. Proof: . g1 is an epimorphism.6 A sequence ( nite or in nite) of groups and homomorphisms 3 2 : : : ! G3 g! G2 g! G1 ! : : : image g3 = ker g2: It is an exact sequence if it is exact at every group. 3. (() If g1 is an epimorphism. g3 is an epimorphism. Then the following are equivalent: 1. Proof: Lemma 6. then 1 Lemma 6. By exactness. Since : G0 ! 0. g1 is a monomorphism. To develop our intuition concerning exact sequences we will prove a few simple lemmas.8 0 ! G1 g! G0 is an exact sequence if and only if g1 is an monomorphism. g2 is the zero homomorphism.

Again. if K := ker g2 . Similarly. @k g be chain complexes.11 Stated as a de nition. So all that remains is to show that the sequence is exact at Ck (X ). Conversely. it appears naturally in many examples.2. image Ik = Ck (E ). E ) ! 0 (6. EXACT SEQUENCES 213 De nition 6. That this is a short exact sequence follows from simple applications of the previous lemmas.8 k 0 ! Ck (E ) I! Ck (X ) is exact. @k g and fCk . Consider a cubical pair (X. then the sequence 2 0 ! G3 ! G2 g! G1 ! 0 is short exact where is the inclusion map. Therefore. in search of the natural de nitions we recall the case of maps between chain 0 0 complexes.2) where Ik is the inclusion map and k is the projection map. Ik is a monomorphism since K(E ) K(X ).6.10 A short exact sequence is an exact sequence of the form 3 2 0 ! G3 g! G2 g! G1 ! 0: Example 6.12 Let 3 2 0 ! G3 g! G2 g! G1 ! 0 be a short exact sequence. Lemma 6. To . Recall that the maps of interest between chain complexes are chain maps F : C ! C 0. Lemma 6. The short exact sequence (6. since Ik is a monomorphism. b b To begin with.2) is called the short exact sequence of a pair. by Lemma 6. i. by de nition of relative chains k is an epimorphism. E ) ! 0 is exact. Proof: We now turn to the question of maps between exact sequences. Similarly. Let fCk . By de nition the kernel of k is Ck (E ). it may appear the a short exact se- quence is a rather obscure notion.7 implies that k Ck (X ) ! Ck (X. Then. Hence. g2 induces an isomorphism from G2 =g3(G3 ) to G1 . E ) and for each k the following sequence k k 0 ! Ck (E ) I! Ck (X ) ! Ck (X. image Ik = ker k .e. However.

@k g.e. In this section we shall prove a theorem that is fundamental to all of homology theory.214 CHAPTER 6. @k g be chain complexes. observe that the two chain complexes and chain map form the following commutative diagram. if Fk is an isomorphism for each k. Let F : A ! B and G : B ! C be chain maps. A homomorphism F from the rst sequence to the second is a collection of group homomorphisms Fk : Gk ! G0k such that the following diagram commutes +1 k : : : ! Gk+1 gk! Gk g! G? 1 ! : : : k ? ? ?Fk+1 ?Fk ?Fk 1 y y y (6. the chain complex in which each group is the trivial group. @k g.3) 0 0 @k+1 0 0 @k 0 : : : ! Ck+1 ! Ck ! Ck 1 ! : : : This leads us to the following de nition for the more restrictive case of exact sequences. i. As a corollary we will answer the motivating question of how relative homology groups are related to the homology groups of the each of spaces in the pair. Let 0 denote the trivial chain complex. B = fBk .3 The Connecting Homomorphism In the previous section we de ned the notion of an exact sequence and proved some simple lemmas. HOMOLOGICAL ALGEBRA begin to view this in the context of exact sequences. and C = fCk .14 Let A = fAk . De nition 6.4) 0 0 gk+1 gk : : : ! G0k+1 ! G0k ! G0k 1 ! : : : F is an isomorphism.13 Let +1 k : : : ! Gk+1 gk! Gk g! Gk 1 ! : : : and 0 g0 +1 k : : : ! G0k+1 k! G0k g! G0k 1 ! : : : be exact sequences. +1 k : : : ! Ck+1 @k! Ck @! Ck 1 ! : : : ? ? ? ?Fk+1 ?Fk ?Fk 1 y y y (6. A B C De nition 6. 6. The sequence 0!A F B G C!0 ! ! .

Proof: 6.) ! (X. Proof: Corollary 6. . Then there exists a long exact sequence : : : ! Hk (E ) I! Hk (X ) ! Hk (X.4 Relative Lefschetz Theorem 6. E ) are inclusion maps.6.4.5 Mayer-Vietoris Sequence .16 (The exact homology sequence of a pair) Let (X. 0!A F B G C!0 ! ! be a short exact sequence of chain complexes. for each k there exist a map @ : Hk (C ) ! Hk 1(A) such that Theorem 6. Then.15 Let : : : ! Hk (A) F! Hk (B) G Hk (C ) @! Hk 1(A) ! : : : ! is a long exact sequence. E ) be a cubical pair. E ) @! Hk 1(E ) ! : : : where I : E ! X and : (X. RELATIVE LEFSCHETZ THEOREM is a short exact sequence of chain complexes if for every k k 0 ! Ak F! Bk Gk Ck ! 0 ! 215 is a short exact sequence.

HOMOLOGICAL ALGEBRA .216 CHAPTER 6.

A relation on X is a subset R X X . z) 2 R When R is an equivalence relation. Example A. m) j m n is a multiple of 2g. De nition A. y) with x 2 X and y 2 Y . R is re exive. x) 2 R for all x 2 X . z) 2 R implies that (x. 3. (x. y) 2 R. y) 2 R and (y.e.1 m = 2ng. y) 2 R implies that (y. i. The cartesian product of X and Y consists of all ordered pairs (x. m) j n and m share a prime factorg: 3.e. i. y) j x 2 X and y 2 Y g: Let X be any set. m) j 2. Consider a the set of integers Z and let R = f(n.Appendix A Equivalence Relations Let X and Y be sets. It is denoted by X Y := f(x. (x. the standard convention is to write x y if and only if (x. 2. 1. i.2 R is an equivalence relation on X if 1. (x. Consider the set of positive integers Z+ and let R = f(n. R is symmetric. x) 2 R. 217 . R is transitive.e. Consider the set of integers Z and let R = f(n.

Then x] = y]. Similarly. It is easy to check that the equivalence classes are disjoint. Finally. x y and hence x] = y]. Observe that if n m then m n is divisible by 2. EQUIVALENCE RELATIONS relation. since any element of E is an equivalence class which can be represented by x] and therefore. . A nal important comment concerning equivalence relations has to do with the functions they induce.3 is an equivalence k n = k m + m n = 2i + 2j = 2(i + j ).1. is a function. To be more precise. z 2 y] means that z y. if n m and m k then there exist integers i and j such that m n = 2i and k m = 2j . Assume that there exists z 2 Z such that z 2 x] and z 2 y]. To see this we must check the three conditions. and hence n k. But this means that n m is divisible by two and so m n. for every x 2 X de ne the equivalence class of x to be the subset x] := fy 2 X j x yg: Because. Let X be a set with an equivalence relation . (x) = x]. For every integer n 2 Z . Let E denote the set of equivalence classes. Furthermore. But this implies that Example A. By de nition z 2 x] means that z x.3 The relation R de ned by Example A. Namely. By transitivity and symmetry. Let x] and y] be equivalence classes. an eqivalence relation is re exive it is clear that x 2 x]. Since equivalence classes are disjoint. n n since n n = 0 which is a multiple of 2. Given an equivalence relation on a set X there is a natural way to partition X into disjoint subsets. Another way of saying this is that if x] 6= y] then x 6 y.218 APPENDIX A. Another standard notation for the set E is X= . is surjective. Let : X ! E be given by (x) = x].

Herlihy and N. Munkres. New Jersey 1975. Inc. 219 . Elements of Algebraic Topology Addison-Wesley. 1984. Prentice-Hall.Bibliography 1] M. Munkres. The topological structure of asynchronous computability. Topology: a rst course. preprint. 3] J. R. Englewood Cli s. 2] J. Shavit. R.

Valette a. Introduction to the Baum-Connes Conjecture

Nekrashevich v. Self-Similar Groups

- Read and print without ads
- Download to keep your version
- Edit, email or read offline

Salmon George

Carl B. Boyer, Uta C. Merzbach - A History of Mathematics

Chow B., Lu P., Ni L. - Hamilton's Ricci Flow, Vol.1(2005)(348)

Advanced Probability Theory for Bio Medical Engineers - John D. Enderle

Statistics Guide

Coherent Stress Testing

Ebooksclub.org a rial Approach to Matrix Theory and Its Applications Discrete Mathematics and Its Applications

Mth 133 Trigonometry

3642125972

Notes - NMA

Function Points Analysis

Geometric Group Theory, Volume 1

Bierens H. - Introduction to the Math. and Stat. Foundations of Eco No Metrics

Exact Solutions and Invariant Subspaces

Quran & Primary Numbers

Calculus

An Episodic History of Mathematics

0521850126.Cambridge.University.Press.The.Mathematics.of.Behavior.Oct.2006

Robust Statistics Theory And Methods

A Logic of Exceptions Thomas Cool Book

Math All

Probability and Statistics Basic a Full Course

Galois

From Determinant to Tensor_bw

Discrete Mathematics 2005

Riemann Geometry 1

SPSS Forecasting 17.0

A Problem Course in Mathematical Logic

Introduction to LATEX

Real-Life Math~Everyday Use of Mathematical Concepts [2002]

Are you sure?

This action might not be possible to undo. Are you sure you want to continue?

CANCEL

OK

You've been reading!

NO, THANKS

OK

scribd