P. 1


|Views: 66|Likes:
Published by Debabrata Bal

More info:

Published by: Debabrata Bal on Apr 19, 2013
Copyright:Attribution Non-commercial


Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See more
See less





First Edition, 2012

ISBN 978-81-323-2072-2

© All rights reserved.

Published by: Library Press 4735/22 Prakashdeep Bldg, Ansari Road, Darya Ganj, Delhi - 110002 Email: info@wtbooks.com 

Table of Contents
Chapter 1 - Introduction to Function Chapter 2 - Inverse Function Chapter 3 - Special Functions & Implicit and Explicit Functions Chapter 4 - Function Composition Chapter 5 - Continuous Function Chapter 6 - Additive Function Chapter 7 - Algebraic Function Chapter 8 - Analytic Function Chapter 9 - Completely Multiplicative Function and Concave Function Chapter 10 - Convex Function Chapter 11 - Differentiable Function Chapter 12 - Elementary Function and Entire Function Chapter 13 - Even and Odd Functions Chapter 14 - Harmonic Function Chapter 15 - Holomorphic Function Chapter 16 - Homogeneous Function Chapter 17 - Indicator Function Chapter 18 - Injective Function Chapter 19 - Measurable Function

An example of a function with the real numbers as both its domain and codomain is the function f(x) = . also known as the input) completely determines another quantity (the value. The argument and the value may be real numbers. Both the domain and the range in the picture are the set of real numbers between -1 and 1. The mathematical concept of a function expresses the intuitive idea that one quantity (the argument of the function. but they can also be elements from any given sets: the domain and the codomain of the function.Chapter 1 Introduction to Function Graph of example function. or the output). A function assigns a unique value to each input of a specified type.5.

many traditions have grown up around their use. There are many ways to describe or represent functions: by a formula. An important operation on functions. and returns an output ƒ(x). which assigns to every real number the real number with twice its value. in such disciplines as real analysis and complex analysis. physics. A function may also be described through its relationship to other functions. . chemistry. and other sciences. by the letter t.3} associates A with 1. In this case. which distinguishes them from numbers. are called function spaces and are studied as objects in their own right.2.2x. by an algorithm that computes it. x. The function itself is most often called f. if the input is a particular time. A table of values is a common way to specify a function in statistics. An example of a function with domain {A. Collections of functions with certain properties. functions include maps between algebraic structures like groups and maps between geometric objects like manifolds. The symbol for the output is called the dependent variable or value and is often represented by the letter y. B with 2. by a plot or a graph. most of which cannot be expressed with a formula or an algorithm. The symbol for the input to a function is often called the independent variable or argument and is often represented by the letter x or. In addition to elementary functions on numbers. In a setting where they have numerical outputs. A function ƒ takes an input. for example. Overview Because functions are so widely used. functions may be added and multiplied. usually required to be closed under certain operations. such as continuous functions and differentiable functions.C} and codomain {1. it is written that f(5) = 10. and thus the notation y = f(x) indicates that a function named f has an input named x and an output named y. yielding new functions. There are uncountably many different functions from the set of natural numbers to itself. a function is a relation between the domain and the codomain that associates each element in the domain with exactly one element in the codomain. In the abstract set-theoretic approach. is the composition of functions. as the inverse function or a solution of a differential equation. One metaphor describes the function as a "machine" or "black box" that converts the input into the output.B. and C with 3.

4) then gives the value 7 associated with the pair (3. Sometimes. in some cases it may be impossible to define such a rule. called the codomain of a function. Y is the codomain. A function of two or more variables is considered in formal mathematics as having a domain consisting of ordered pairs or tuples of the argument values. especially in computer science. For example. and an association between the two that can be described by a set of ordered pairs like ((3. or as a function f(n) = 1/n from the set of natural numbers into the set of rational numbers. Sum then has a domain consisting of elements like (3.y) = x+y operating on integers is the function Sum with a domain consisting of pairs of integers.. and the parts are called the fibers or level sets of the function at each element of the codomain.The set of all permitted inputs to a given function is called the domain of the function. . Furthermore. For example. Definition One precise definition of a function is that it consists of an ordered triple of sets. Dually. when a name for the function is not needed. as its image the set of all non-negative real numbers. can be written as the ordered sequence <1/n> where n is a natural number. although each input element is still associated to one and only one output. a surjective function partitions its domain into disjoint sets indexed by the codomain.. Y. (A non-surjective function divides its domain into disjoint and possibly-empty subsets). This partition is known as the kernel of the function. so care needs to be taken when using the word. implies ƒ(3) = 7. 1/2. It is usual practice in mathematics to introduce functions with temporary names like ƒ. F). the function f(x) = x2 could take as its domain the set of all real numbers. For example Sum(x.4). a codomain of integers. The set of all resulting outputs is called the image or range of the function... . and F is a . Functions need not act on numbers: the domain and codomain of a function may be arbitrary sets. Evaluating Sum(3. the association between inputs and outputs in a choice function often lacks any fixed rule. functions need not be described by any expression. 1/3. the sequence 1.4). which may be written as (X. for example. In that case. ƒ(x) = 2x+1. for example. 7). we would describe f as a real-valued function of a real variable. 1/n. Thus. X is the domain of the function. One example of a function that acts on non-numeric inputs takes English words as inputs and returns the first letter of the input word as output. rule or algorithm: indeed. the form y = 2x+1 may be used. the term "range" refers to the codomain rather than the image. it may be given a more permanent name as. For example.. and as its codomain the set of all real numbers. If a function is often used. The image is often a subset of some larger set.4). A family of objects indexed by a set is equivalent to a function.

with the restriction that the graph should not contain two distinct ordered pairs with the same first element. Indeed. Some authors (especially in set theory) define a function as simply its graph f. given such a graph. the graph contains at most one ordered pair (and possibly none) with it as a first element. The notation ƒ:X→Y indicates that ƒ is a function with domain X and codomain Y.set of ordered pairs. others to mean the codomain. Some authors use the term "range" to mean the image. b). i. and so is only a partial function from the real line to the real line. (3. In other parts of mathematics. and only the relationship between the input and output is given. A relation that is both left-total and single-valued is a function. the relation. including recursion theory and functional analysis. it is convenient to study partial functions in which some values of the domain have no association in the graph. A function is a special case of a more general mathematical concept. In each of these ordered pairs (a. Thus is usually written as The graph of a function is its set of ordered pairs. A relation is "single-valued" or "functional" when for each element of the domain set. one can construct a suitable triple by taking the set of all first elements as the domain and the set of all second elements as the codomain: this automatically causes the function to be total and surjective . In some parts of mathematics. the second element b is from the codomain. Such a set can be plotted on a pair of coordinate axes. with the corresponding term singlevalued function for ordinary functions. the graph contains at least one ordered pair with it as a first element (and possibly more than one). most authors in . However.. non-single-valued relations are similarly conflated with functions: these are called multivalued functions. for which the restriction that each element of the domain appear as the first element in one and only one ordered pair is removed (or. the restriction that each input be associated to exactly one output). the function f such that f(x) = 1/x does not define a value for x = 0. for example. 9) is the point of intersection of the lines x = 3 and y = 9. the first element a is from the domain. and every element in the domain is the first element in one and only one ordered pair. A relation is called "left-total" or simply "total" when for each element of the domain. The term total function can be used to stress the fact that every element of the domain does appear as the first element of an ordered pair in the graph. the domain and codomain are understood from context. single-valued relations. In most practical situations.e. For example. The set of all b is known as the image of the function. in other words.

an example being where the first part is read: • • • "ƒ is a function from N to R" (one often writes informally "Let ƒ: X → Y" to mean "Let ƒ be a function from X to Y"). with the ordered pairs being the Cartesian coordinates of points. Many operations in set theory—such as the power set—have the class of all sets as their domain. or the image of x under ƒ. and the second part is read: • maps to . use the terms "function" and "map" to refer to different types of functions. For each argument value x. or "ƒ is a function on N into R". the real numbers. Other specific types of functions include functionals and operators. ƒ(x)). A function can also be called a map or a mapping. or "ƒ is an R-valued function of an N-valued variable". however. The graph of a function ƒ is the set of all ordered pairs (x. If X and Y are subsets of R. Notation Formal description of a function typically involves the function's name. output of ƒ for an argument x. Vocabulary A specific input in a function is called an argument of the function. its domain. then this definition coincides with the familiar sense of "graph" as a picture or plot of the function. Some authors. therefore. they do not fit the set-theoretical definition above outlined. although they are informally described as functions. and a rule of correspondence. for all x in the domain X. the corresponding unique y in the codomain is called the function value at x. The image of x may be written as ƒ(x) or as y.advanced mathematics outside of set theory prefer the greater power of expression afforded by defining a function as an ordered triple of sets. its codomain. Thus we frequently see a two-part notation.

For example. thus: sin x. for example.y)). This intuitive concept is formalized by a function whose domain is the Cartesian product of two or more sets.y.y) a function of two variables. y) = x·y. x·y). while the second component is a single integer. codomain Z. and then . the gamma function. There is some loss of information: we no longer are explicitly given the domain N and codomain R. The function value of the pair (x. and. However. this long form might be abbreviated where f(n) is read as "f as function of n" or "f of n". Currying An alternative approach to handling functions with multiple arguments is to transform them into a chain of functions that each takes a single argument. consider the function that associates two integers to their product: ƒ(x. For instance. The concept can still further be extended by considering a function that also produces output that is expressed as several variables. consider the function swap(x. x and y. is known as postfix notation. the set of all pairs ((x. x) with domain R×R and codomain R×R as well. The pair (y. Writing the function after its argument. and maps n to itself divided by π. the real numbers as codomain. Less formally.f(x. for graph. For example. Functions of two variables may be plotted on the three-dimensional Cartesian as ordered triples of the form (x. the factorial function is customarily written n!. even though its generalization. y) = (y. is written Γ(n).y) is ƒ((x. Functions with multiple inputs and outputs The concept of function can be extended to an object that takes a combination of two (or more) argument values to a single result.Here the function named "ƒ" has the natural numbers as domain. This function can be defined formally as having domain Z×Z. one can interpret Add(3.y)).y). though in some formal settings the consistent use of either prefix or postfix notation eliminates the need for any parentheses. Parentheses are still used to resolve ambiguities and denote precedence. this is known as prefix notation. it is customary to drop one set of parentheses and consider ƒ(x. x) is a single value in the codomain seen as a Cartesian product.5) to mean "first produce a function that adds 3 to its argument. the set of all integer pairs. Note that the first component of any such pair is itself a pair (of integers). as in x ƒ. It is common to omit the parentheses around the argument when there is little chance of confusion.

and the bijections. since the codomain is not defined. which are both one-to-one and onto. y) and ×(x. which have the property that for every y in the codomain there is an x in the domain such that ƒ(x) = y. For example. Injective and surjective functions Three important kinds of function are the injections (or one-to-one functions). can be viewed as functions from R×R to R. y). This transformation is called currying: Add 3 is curry(Add) applied to 3. Binary operations The familiar binary operations of arithmetic. This nomenclature was introduced by the Bourbaki group. For example. . where n-ary functions are used to model the operations of arbitrary algebraic structures. which have the property that if ƒ(a) = ƒ(b) then a must equal b. There is a bijection between the function spaces CA×B and (CB)A. since juxtaposition of multiple arguments—as in (ƒ x y)—naturally maps to evaluation of a curried function.apply the 'Add 3' function to 5". When the definition of a function by its graph only is used. addition and multiplication are written in the infix notation: x+y and x×y instead of +(x. the surjections (or onto functions). the "surjection" must be accompanied with a statement about the set the function maps onto. This view is generalized in abstract algebra. Traditionally. addition and multiplication. we might say ƒ maps onto the set of all real numbers. When working with curried functions it is customary to use prefix notation with function application considered left-associative. an abstract group is defined as a set X and a function ƒ from X×X to X that satisfies certain properties.

We remember the order by reading the notation as "g of ƒ". suppose ƒ(x) = x2 and g(x) = x+1. reversing English reading order. For example. The first takes input x and outputs f(x).Function composition A composite function g(f(x)) can be visualized as the combination of two "machines". Then g(ƒ(x)) = x2+1. which is x2+2x+1. because rarely do we get the same result both ways. . while ƒ(g(x)) = (x+1)2. The functions ƒ: X → Y and g: Y → Z can be composed by first applying ƒ to an argument x to obtain y = ƒ(x) and then applying g to y to obtain z = g(y). The composite function formed in this way from general ƒ and g may be written This notation follows the form such that The function on the right acts first and the function on the left acts second. a different function. The order is important. The function composition of two or more functions takes the output of one or more functions as the input of others. The second takes f(x) and outputs g(f(x)).

If A is any subset of the domain. Identity function The unique function over a set X that maps each element to itself is called the identity function for X. The overriding of f: X → Y by g: W → Y (also called overriding union) is an extension of g denoted as (f ⊕ g): (X ∪ W) → Y. Overriding is an associative operation. an identity function is "neutral": if ƒ is any function from X to Y. then the union of f and g is defined and is equal to their overriding union. If g is a restriction of ƒ. More precisely. writes the function name above the arrow. We say the ƒ(A) is the image of A under f. and multiplication of real numbers. where g produces the complex number eix.g. If f|X ∩ W and g|X ∩ W are pointwise equal (e. Its graph is the set-theoretical union of the graphs of g and f|X \ W. we may write A more elaborate form of this is the commutative diagram. if ƒ is followed by g. a restriction of a function ƒ is the result of trimming its domain. An alternative to the colon notation. namely the addition. Thus. Each set has its own identity function. the function given above by the formula y = 5x−20x3+16x5 can be obtained by composing several functions. then ƒ(A) is the subset of im ƒ consisting of all images of elements of A. Under composition. then it is said that ƒ is an extension of g. . then Restrictions and extensions Informally.In a similar way. and S is any subset of X. the domains of f and g are disjoint). it relates any element of the domain of g to its image under g. and any other element of the domain of f to its image under f. Image of a set The concept of the image can be extended from the image of a point to the image of a set. the restriction of ƒ to S is the function ƒ|S from S to Y such that ƒ|S(s) = ƒ(s) for all s in S. This definition agrees with the definition of union for binary relations. For example. if ƒ is a function from a X to Y. convenient when functions are being composed. it has the empty function as an identity element. negation. so the subscript cannot be omitted unless the set can be inferred from context. and typically denoted by idX..

Notice that the image of ƒ is the image ƒ(X) of its domain. limits. complete inverse image) of a subset B of the codomain Y under a function ƒ is the subset of the domain X defined by So. The usual convention about the preimage of an element is that ƒ−1(b) means ƒ−1({b}). 9} under the squaring function is the set {−3. The term kernel can refer to a number of related concepts. Inverse image The inverse image (or preimage. In some fields (e.Use of ƒ(A) to denote the image of a subset A⊆X is consistent so long as no subset of the domain is also an element of the domain. if ƒ(x) = 7. The preimage of a singleton set is sometimes called a fiber. some authors write ƒ`x instead of ƒ(x). for example. a function ƒ may be defined by simply tabulating all the arguments x and their corresponding function values ƒ(x). and that the image of ƒ is a subset of its codomain. algebraic or analytic closure. where ordinals are also sets of ordinals) it is convenient or even necessary to distinguish the two concepts. infinite series. the preimage of a singleton set (a set with exactly one element) may contain any number of elements. the customary notation is ƒ[A] for the set { ƒ(x): x ∈ A }. or (more generally) an algorithm — a recipe that tells how to compute the value of ƒ(x) given any x in the domain. some authors use square brackets to avoid confusion between the inverse image and the inverse function. Specifying a function A function can be defined by any mathematical condition relating each argument to the corresponding output value. and as solutions to integral and differential equations. Thus the preimage of an element in the codomain is a subset of the domain. and ƒ``A instead of ƒ[A]. analytic continuation. or more precisely. then the preimage of {5} is the empty set but the preimage of {7} is the entire domain.−2.. a function is defined by a formula. Thus they would write ƒ−1[B] and ƒ−1[b] for the preimage of a set and a singleton. There are many other ways of defining functions. induction or recursion.2. in set theory. The .g.3}. i. For example. More commonly. the preimage of {4. Examples include piecewise definitions.e In the same way as for the image. In general. If the domain is finite.

used for the set of permutations where |X!| = |X|!. it may reasonably be concluded that ƒ ∈ [X → Y]. including the busy beaver function and functions related to the halting problem and other undecidable problems. or by YX. Many of the functions studied in the context of number theory are computable. and the binomial coefficient sign where If ƒ: X → Y. the factorial sign X!. . because the number of possible algorithms is. Computability Functions that send integers to integers. can sometimes be defined by an algorithm. Function spaces The set of all functions from a set X to a set Y is denoted by X → Y. where |X×Y| = |X|·|Y|. then the sum function ƒ + g: X → R and the product function ƒ ⋅ g: X → R can be defined as follows: . Functions definable by an algorithm are called computable functions. in the sense of cardinality. The latter notation is motivated by the fact that. then the number of functions X → Y is |YX| = |Y||X|. Fundamental results of computability theory show that there are functions that can be precisely defined but are not computable. Moreover. Other examples are the multiplication sign X×Y used for the Cartesian product. This is an example of the convention from enumerative combinatorics that provides notations for sets based on their cardinalities. the Euclidean algorithm gives a precise process to compute the greatest common divisor of two positive integers. For example. which gives a precise description of a set of steps for computing the output of the function from its input.lambda calculus provides a powerful and flexible syntax for defining and combining functions of several variables. or finite strings to finite strings. almost all functions from the integers to integers are not computable. Specific examples of uncomputable functions are known. Thus most functions from integers to integers are not computable. by [X → Y]. The number of computable functions from integers to integers is countable. The number of all functions from integers to integers is higher: the same as the cardinality of the real numbers. used for the set of n-element subsets Pointwise operations If ƒ: X → R and g: X → R are functions with a common domain of X and common codomain of a ring R. when X and Y are finite and of size |X| and |Y|.

unimodal holomorphic. some general ideas about independent and dependent variable quantities seem to be present. This is an example of climbing up in abstraction. By taking some other algebraic structure A in the place of R. and as codomain functions. entire vector-valued computable History Functions prior to Leibniz Historically. . transcendental trigonometric fractal odd or even convex. or particular applications. rational algebraic. . Here is a partial list: • • • • • • • • • • • • bijection. Other properties There are many other special classes of functions that are important to particular branches of mathematics. or singularly: o injective. injection and surjection. This turns the set of all such functions into a ring. to functions of more complex types. we can turn the set of all functions from X to A into an algebraic structure of the same type in an analogous way. meromorphic. .for all x in X. The binary operations in that ring have as domain ordered pairs of functions. polynomial. Among them is Oresme (1323-1382) . o surjective. integrable linear. monotonic. In his theory. some mathematicians can be regarded as having foreseen and come close to a modern formulation of the concept of function. Ponte further notes that "The emergence of a notion of function as an individualized mathematical entity can be traced to the beginnings of infinitesimal calculus". and o bijective function continuous differentiable.

to describe a quantity related to a curve. But there is ample evidence that he had no idea of this concept. However. x2+3x+2. but Dirichlet's claim to this formalization is disputed by Imre Lakatos: There is no such definition in Dirichlet's works at all. These functions were first thought to be only theoretical curiosities. both are measurements of the output or the change in the output as it depends on the input or the change in the input. Alexis Claude Clairaut (in approximately 1734) and Euler introduced the familiar notation " f(x) ". in a precise sense. Eves asserts that "the student of mathematics usually meets the Dirichlet definition of function in his introductory course in calculus. (Proofs and Refutations. Johann Bernoulli "by 1718. Such functions have since been applied to the modeling of physical phenomena such as Brownian motion. 151. During the 19th century. had come to regard a function as any expression made up of a variable and some constants".. something no mathematician would claim today..The notion of "function" in analysis As a mathematical term. In his [1837]. and Leonhard Euler during the mid-18th century used the word to describe an expression or formula involving variables and constants e. mathematicians started to formalize all the different branches of mathematics.) In the context of "the Differential Calculus" George Boole defined (circa 1849) the notion of a function as follows: . when he discusses piecewise continuous functions. claimed that every function had a Fourier series. for example. more common than differentiable functions. By broadening the definition of functions. For this type of function. mathematicians were able to study "strange" mathematical objects such as continuous functions that are nowhere differentiable. At first. Joseph Fourier. The functions Leibniz considered are today called differentiable functions.g. Weierstrass advocated building calculus on arithmetic rather than on geometry. the idea of a function was rather limited. in a 1673 letter. Such functions are the basis of calculus. Cambridge University Press 1976. Dirichlet and Lobachevsky are traditionally credited with independently giving the modern "formal" definition of a function as a relation in which every first element has a unique second element. which favoured Euler's definition over Leibniz's. such as a curve's slope at a specific point.. and they were collectively called "monsters" as late as the turn of the 20th century. one can talk about limits and derivatives. "function" was coined by Gottfried Leibniz. he says that at points of discontinuity the function has two values: . powerful techniques from functional analysis have shown that these functions are. for instance.

and not upon the particular matters spoken of". the subject. and lastly a new symbolism for logical analysis of the notion of "relation" (he uses the word with respect to this example " X)Y " (p. the notion of "abstraction" is there. At this time the notion of (logical) "function" is not explicit." The logician's "function" prior to 1850 Logicians of this time were primarily involved with analyzing syllogisms (the 2000 yearold Aristotelian forms and otherwise). 75)) is there: " A1 X)Y To take an X it is necessary to take a Y" [or To be an X it is necessary to be a Y] " A1 Y)X To take an Y it is sufficient to take a X" [or To be a Y it is sufficient to be an X]. . or as Augustus De Morgan (1847) stated it: "the examination of that part of reasoning which depends upon the manner in which inferences are formed. "is". copula. That quantity whose variation is referred to the variation of the former is said to be a function of it. we might represent the aggregate class oxen and horses by x + y". Necessary and Probable" observes that "[a] logical truth depends upon the structure of the statement. etc. the copula is made as absract as the terms". The Calculus of Inference. is in a more especial sense the science of reasoning by signs". and the investigation of general maxims and rules for constructing arguments". . and predicate. While the word "function" does not appear. . In his 1848 The Nature of Logic Boole asserts that "logic . "variables" are there. the introduction of a symbolic algebra with respect to these variables. . Like De Morgan he uses the notion of "variable" drawn from analysis. the introduction of variables. The Differential calculus enables us in every case to pass from the function to the limit. but at least in the work of De Morgan and George Boole it is implied: we see abstraction of the argument forms. is called the independent variable. He immediately (p. he gives an example of "represent[ing] the class oxen by x and that of horses by y and the conjunction and by the sign + . and Y represent. he wastes no time (preface page i) abstracting: "In the form of the proposition. respectively. the notion of inclusion in his symbolism “all of the Δ is in the О” (p. But in the very Idea of an Operation is . . 1) casts what he calls "the proposition" (present-day propositional function or relation) into a form such as "X is Y"."That quantity whose variation is uniform . and he briefly discusses the notions of "belonging to" and "class": "An individual may possess a great variety of attributes and thus belonging to a great variety of different classes" . De Morgan's 1847 "FORMAL LOGIC OR. and some of the notions of set theory. . . 9) is there. the idea of an inverse operation. . . This it does by a certain Operation. To effect that inverse operation in the present instance is the business of the Int[egral] Calculus. where the symbols X.

Any algebraic expression involving symbol x is termed a function of x. cf page xxi) plus the circle-diagrams historically associated with Venn to describe "class relations".g. to show that all mathematics follows from symbolic logic. as far as possible. the bar over a variable to indicate not-x (page 43). and "propositional function" (all on p. xy is the logical AND(x. the Logicists.. e. but also as a reaction against Russell's proposed solution. But by the late 19th century the logicians' research into the foundations of mathematics was undergoing a major split. from a foundation in the logic of propositions and propositional functions". f(x) never stands for anything but a logical class. x + y is the logical OR(x. and "the special law" xx = x2 = x. "the relation of inclusion and exclusion of two classes to one another". x(x+y) is xx+xy. In his 1881 Symbolic Logic Venn was using the words "logical function" and the contemporary symbolism (x = f(y). their difference declared equal to zero. John Venn's Symbolic Logic 1881 In his An Investigation into the laws of thought Boole now defined a function in terms of a symbol x as follows: "8. it may be a class indicated by certain inverse logical operations. and secondly to discover. here too the notion of "propositional function" plays a role. 1−x is logical NOT(x). can probably be summed up best by Bertrand Russell 1903:9 -.y). emerged with Georg Cantor's "set theory" (1870–1890) but were driven forward partly as a result of Russell's discovery of a paradox that could be derived from Frege's conception of "function"."to fulfil two objects. f(x) with us will never be anything else than a general expression for such logical classes of things as may fairly find a place in ordinary Logic".The logicians' "function" 1850-1950 Eves observes "that logicians have endeavored to push down further the starting level of the definitional development of mathematics and to derive the theory of sets... that is.the first axiomatic set theory. Indeed he equated unequivocally the notion of "logical function" with "class" [modern "set"]: ". George Boole's The Laws of Thought 1854. first. 10). But however composed or derived. the set-theorists. y = f−1(x). it may be composed of two groups of classes equal to one another. a logical equation. the notions "'quantifying' our predicate". Definition. what are the principles of symbolic logic itself. etc. It may be a compound class aggregated of many simple classes. y). "propositions in respect of their extension". The direction of the first group.-. or classes." The second group of logicians. . or what is the same thing. on the view adopted in this book. and may be represented by the abbreviated form f(x)" Boole then used algebraic expressions to define both algebraic and logical notions. Zermelo's set-theoretic response was his 1908 Investigations in the foundations of set theory I -.

Finally. which he believes "will stand the test of time.Frege's Begriffsschrift 1879 Gottlob Frege's Begriffsschrift (1879) preceded Giuseppe Peano (1889).." "Nitrogen is lighter than carbon dioxide. There are three statements: • • • "Hydrogen is lighter than carbon dioxide. remove "carbon dioxide" to yield the invariant part (the function) as: • "." "Oxygen is lighter than carbon dioxide. B) differs from Ψ(B. oxygen. this makes a second statement. A)". . whereas the two-argument function he symbolizes as Ψ(A. At the outset Frege abandons the traditional "concepts subject and predicate". not. Frege allows for the case of two (or more arguments). Russell in turn influenced much of 20th-century mathematics and logic through his Principia Mathematica (1913) jointly authored with Alfred North Whitehead. deserves attention"." Now observe in all three a "stable component..g. the demonstration of the connection between the meanings of the words if. all. " The one-argument function Frege generalizes into the form Φ(A) where A is the argument and Φ( ) represents the function. there is. and so forth. or. Both writers strongly influenced Bertrand Russell (1903). the word "hydrogen") and replace it with the sign for oxygen (i.. He notes that we could have derived the function as "Hydrogen is lighter than .. Now remove the sign for hydrogen (i.. some. Furthermore. For example.. but Peano had no knowledge of Frege 1879 until after he had published his 1889. hydrogen. replacing them with argument and function respectively.e. with an argument position on the right. B) with A and B the arguments and Ψ( .. the exact observation is made by Peano. Do this again (using either statement) and substitute the sign for nitrogen (i. It is easy to see how regarding a content as a function of an argument leads to the formation of concepts.e. or nitrogen]. the word "oxygen"). the word "nitrogen") and note that "This changes the meaning in such a way that "oxygen" or "nitrogen" enters into the relations in which "hydrogen" stood before".. Frege calls the argument of the function "[t]he sign [e. "." as well.. is lighter than carbon dioxide". representing the totality of [the] relations". . call this the function.. is the function. regarded as replaceable by others that denotes the object standing in these relations".e... and. Frege begins his discussion of "function" with an example: Begin with the expression "Hydrogen is lighter than carbon dioxide". Using his unique symbolism he translates for the reader the following symbolism: . ) the function and cautions that "in general Ψ(A. is lighter than . i.e.

φ is a "function presign". or aggregate of objects". we assume it is possible to deduce φx = φy".Ψ(A. (a = b) = (b = a). in the above instance 2*( )3 + ( ). but also by the fact that they contain variables.e. For example if φ is the function presign a+. these generalizations Russell named propositional functions". Mathematical propositions are not only characterized by the fact that they assert implications. and these words are the marks of a variable and a formal implication". IF ((a = b) AND (b = c)) THEN (a = c). He then introduces φ. Russell agreed with Frege's notion of "function" in one sense: "He regards functions -. in Appendix A "The Logical and Arithmetical Doctrines of Frege" of The Principles of Mathematics. the objects of which satisfy three simple equality-conditions. and so long as this is possible. Russell arrives at a discussion of Frege's notion of function. "a sign or an aggregate of signs such that if x is an object of the class s. then φx yields a+x. Indeed he cites and quotes from Frege's Begriffsschrift and presents a vivid example from Frege's 1891 Function und Begriff: That "the essence of the arithmetical function 2*x3+x is what is left when the x is taken away. B) can be translated by "B stands in the relation Ψ to A" or "B is a result of an application of the procedure Ψ to the object A". that the words any or some occur. First Peano defines the sign "K means class. and requires careful examination". Bertrand Russell's The Principles of Mathematics 1903 While the influence of Cantor and Peano was paramount. in all mathematical propositions.. or if φ is the function postsign +a then xφ yields x+a. our proposition can be generalized. ". ."We can read |--. it is the business of mathematics to do it". and gives us.. |--.and in this I agree with him -. The notion of the variable is one of the most difficult with which logic has to deal. i. the formal essence of a proposition . We shall find always. For the present.. Peano adds two conditions on these new objects: First. Given all these conditions are met. even where at first sight they might seem to be absent.Φ(A) as "A has the property Φ. For Russell the bedeviling notion is that of "variable": "6. The argument x does not belong to the function but the two taken together make the whole". As expressed by Russell "the process of transforming constants in a proposition into variables leads to what is called generalization. Peano 1889 The Principles of Arithmetic 1889 Peano defined the notion of "function" in a manner somewhat similar to Frege. Likewise he identifies a "function postsign". secondly.. a = a. . So long as any term in our proposition can be turned into a variable. that "if x and y are objects of class s and if x = y. that the three equality-conditions hold for the objects φx.as more fundamental than predicates and . as it were.. . In response to his 1902 exchange of letters with Frege about the contradiction he discovered in Frege's Begriffsschrift Russell tacked this section on at the last moment.a point in which Frege's work is very important. I openly wish to make it plain that there are variables in all mathematical propositions. but without the precision. the expression φx denotes a new object".

in particular "he thinks that. . this proposition is called a "value" of the propositional function. 23). considered the propositional function fundamental: "Propositional functions are the fundamental kind from which the more usual kinds of function. Substitution of one of these values for variable ŷ yields a proposition. given the four values "Bob". 15). To . Russell defines functions of propositions with arguments." A proposition.. When one goes to form the class α: φŷ: "ŷ is hurt". "a class [set] is all objects satisfying some propositional function" (p. Finally. like Frege. Evolution of Russell's notion of "function" 1908-1913 Russell would carry his ideas forward in his 1908 Mathematical logical as based on the theory of types and into his and Whitehead's 1910-1913 Principia Mathematica. per Russell's definition.relations" but Russell rejected Frege's "theory of subject and assertion". and we can only assign "significant" propositions to the variables p and q). If a proposition's truth value is "truth" then the variable's value is said to satisfy the propositional function.g.this is how the contemporary notions of "For all ∀" and "there exists at least one instance ∃" enter the treatment (p. suppose one were to form the "function of propositions with arguments" p1: "NOT(p) AND q" and assign its variables the values of p: "Bob is hurt" and q: "This bird is hurt". "Emily the rabbit is hurt" and "y is hurt. as φŷ: "ŷ is hurt". The functions of propositions . An example may help. . the reader may be confused by Russell's "propositional function". we will assign just 4 values to the variable ŷ: "Bob".. if a term a occurs in a proposition. ambiguous.. and "y". By the time of Principia Mathematica Russell. only "This bird" is included. truth. Then the "function of propositions with arguments" is p1: NOT("Bob is hurt") AND "This bird is hurt"). only the value "This bird" of the variable ŷ satisfies' the propositional function φŷ: "ŷ is hurt". "Emily the rabbit". Russell writes a propositional function in its raw form. are a particular case of propositional functions". . and "y is hurt" is ambiguous as to its truth value because the argument y itself is ambiguous. the proposition can always be analysed into a and an assertion about a". Propositional functions: Because his terminology is different from the contemporary. "Bob is hurt". e. if its truth is determinate—has a truth-value of truth or falsity. are called “descriptive functions". such as “sin ‘’x’’ or log x or "the father of x" are derived. For example. In our example there are four values of the propositional function. indeterminate. (We are restricted to the logical linkages NOT. While the two propositions "Bob is hurt" and "This bird is hurt" are significant (both have truth values). "This bird".e. . if it is significant—i. and truth-functions f(p). e. Note the word "all'" -. "This bird is hurt". "This bird". These derivative functions . To continue the example: Suppose (from outside the mathematics/logic) one determines that the propositions "Bob is hurt" has a truth value of "falsity". (Observe the circumflex or "hat" over the variable y). "This bird is hurt" has a truth value of "truth". AND. For our example.g. "Emily the rabbit" and "y" for variable ŷ and their respective truth-values: falsity. OR and IMPLIES. "Emily the rabbit is hurt" has an indeterminate truth value because "Emily the rabbit" doesn't exist.

"effective method" or "algorithm". and "sin" would stand for the relation sin 'y has to y".g.e.e. Thus in our notation "sin y" would be written " sin 'y ". we shall call it a descriptive function. x = sin(y). "relation") φŷ. The reader should be warned here that the order of the variables are reversed! y is the independent variable and x is the dependent variable. In his 1927 The Foundations of Mathematics Hilbert frames the notion of function in terms of the existence of an "object": 13. step-by-step procedure that would succeed in computing . Recursion theory and computability: But the unexpected outcome of Hilbert's and his student Bernays's effort was failure. e. Russell symbolizes the descriptive function as "the object standing in relation to y": R'y =DEF (ιx)(x R y). secondly to represent the "object of which [a proposition] holds". This broad definition of a function encompasses more relations than are ordinarily considered functions in contemporary mathematics. The Formalist's "function": David Hilbert's axiomatization of mathematics (1904-1927) David Hilbert set himself the goal of "formalizing" classical mathematics "as a formal axiomatic theory. pp. which yields a truth value of "truth". and lastly how to cast it into the choice function. f(p1): f(NOT("Bob is hurt") AND "This bird is hurt")).e.] Hilbert then illustrates the three ways how the ε-function is to be used. All the ordinary functions of mathematics are of this kind. firstly as the "for all" and "there exists" notions. an explicit. 26–28 defined a function as a relation between two variables x and y such that "to some values of x at any rate correspond values of y." He neither required the function to be defined for all values of x nor to associate each value of x to a single value of y. but not a propositional function [sic]. At about the same time.. A(a) --> A(ε(A)) Here ε(A) stands for an object of which the proposition A(a) certainly holds if it holds of any object at all. then defines a descriptive function (pages 30ff) as the unique value ιx that satisfies the (2-variable) propositional function (i. i. and this theory shall be proved to be consistent.determine the truth value of this "function of propositions with arguments" we submit it to a "truth function". let us call ε the logical ε-function". The notion of a "many-one" functional relation": Russell first discusses the notion of "identity".. [The arrow indicates “implies”.g.. e.. N. free from contradiction" . Russell repeats that "R'y is a function of y. mathematicians set about to define what was meant by an "effectively calculable function" (Alonzo Church 1936).B. that is.. in an effort to solve Hilbert's Entscheidungsproblem. i. Hardy 1908 Hardy 1908.

" From this unconstrained situation Russell was able to form a paradox: "You state .. too. Church's thesis holds that this class of functions exhausts all the number-theoretic functions that can be calculated by an algorithm. it may also be that the argument is determinate and the function indeterminate . in rapid succession. in Turing's words. including Church's lambda calculus (1936). can act as the indeterminate element. From this point forward development of the foundations of mathematics became an exercise in how to dodge "Russell's paradox". Jevons (1880). 1910–1913) that he proposed as a means to evade the antinomies. Let w be the predicate: to be a predicate that cannot be predicated of itself. consternation. This I formerly believed.. Stephen Kleene's μ-recursive functions(1936) and Allan Turing's (1936-7) notion of replacing human "computers" with utterly-mechanical "computing machines". . by the discovery of more antinomies in the early 20th century (e. I would almost say. Russell's paradox 1902 In 1902 Russell sent a letter to Frege pointing out that Frege's 1879 Begriffsschrift allowed a function to be an argument of itself: "On the other hand.g. and by resistance to Russell's complex treatment of logic and dislike of his axiom of reducibility (1908. the 1897 Burali-Forti paradox and the 1905 Richard paradox). Can w be predicated of itself?" Frege responded promptly that "Your discovery of the contradiction caused me the greatest surprise and.a function.. framed as it was in "the bare [set-theoretic] notions of set and element". .. "there can be no general process for determining whether a given formula U of the functional calculus K [Principia Mathematica] is provable". Frege 1879 and Peano (1889). paradox) in this treatment (Cantor's paradox). It was shown that all of these models could compute the same class of computable functions. since it has shaken the basis on which I intended to build arithmetic". . Development of the set-theoretic definition of "function" Set theory began with the work of the logicians with the notion of "class" (modern "set") for example De Morgan (1847). but now this view seems doubtful to me because of the following contradiction. Venn 1881. that a function. Various models for algorithms appeared. The outcomes of these efforts were vivid demonstrations that. It was given a push by Georg Cantor's attempt to define the infinite in settheoretic treatment(1870–1890) and a subsequent discovery of an antinomy (contradiction. by Russell's discovery (1902) of an antinomy in Frege's 1879 (Russell's paradox).

and "logical calculus. "predicate". and it is Weyl's and Skolem's approach to the formulation of the axiom of separation that is generally adopted. Fraenkel. .Zermelo's set theory (1908) modified by Skolem (1922) The notion of "function" appears as Zermelo's axiom III—the Axiom of Separation (Axiom der Aussonderung). This is the formal advantage that it avoids the necessity for the primitive proposition . B). Whenever the propositional function Φ(x) is definite for all elements of a set M. and is fixed by Weyl. a finite expression constructed from elementary propositions of the form a ε b or a = b by means of the five operations [logical conjunction. . As noted above. The Wiener–Hausdorff–Kuratowski "ordered pair" definition 1914–1921 The history of the notion of "ordered pair" is not clear. . As there is no universal set—sets originate by way of Axiom II from elements of (nonset) domain B -.". Today an axiomatization of set theory is usually embedded in a logical calculus.." This shift in terminology is discussed more in the section that covers "function" in contemporary set theory.. B)) as a "class of couples" but rejected it: "There is a temptation to regard a relation as definable in extension as a class of couples. Russell (1903) considered the definition of a relation (such as Ψ(A.. but rather one sees the words "formula".. universal quantification. In fact Skolem in his 1922 referred to this "definite criterion" or "property" as a "definite proposition": ". M possesses a subset MΦ containing as elements precisely those elements x of M for which Φ(x) is true". disjunction. In this quote the reader may observe a shift in terminology: nowhere is mentioned the notion of "propositional function"... =. This axiom constrains us to use a propositional function Φ(x) to "separate" a subset MΦ from a previously formed set M: "AXIOM III. and existential quantification]. by a well-formed formula in the simple predicate calculus of first order in which the sole predicate constants are ε and possibly. and von Neumann. Skolem. negation.this disposes of the Russell antinomy so far as we are concerned". Norbert Wiener in his 1914 (see below) observes that his own treatment essentially "revert(s) to Schröder's treatment of a relation as a class of ordered couples". "predicate calculus". van Heijenoort summarizes: "A property is definite in Skolem's sense if it is expressed . Frege (1879) proposed an intuitive ordering in his definition of a two-argument function Ψ(A. But Zermelo's "definite criterion" is imprecise. (Axiom of separation).

To quote Wiener the second axiom *12. We shall call it a "couple with sense.. 2} }. A few years later Kuratowski (1921) offered a definition that has been widely used ever since. i. p. y) is true". classes and relations". y) is different from the couple (y.e. descriptions. namely { {a. y) for which φ(x.1.11.. Observe that while Wiener "reduced" the relational *12.1. the relation determined by φ(x. . y) is the class of couples (x. In a footnote he clarified his notion and arrived at this definition: "Such a couple has a sense. using Whitehead and Russell's symbolism. was historically important in reducing the theory of relations to the theory of sets. are to be taken in extension". {b. however. . it may also be called an ordered couple. stating that "mathematics is always concerned with extensions rather than intensions" and "Relations.1}. he and Alfred North Whitehead would carry this treatment over to Principia Mathematica 1910-1913 with a further refinement called "a matrix"..11 "is involved only in the theory of relations". as a class of couples . the purpose of which were to reduce (single-variable) propositional functions and (dual-variable) relations to a "lower" form (and ultimately into a completely extensional form)." By 1910-1913 and Principia Mathematica Russell had given up on the requirement for an intensional definition of a relation. The first axiom is *12. the second is *12. indeed he declared this "essential to the treatment of identity.. . Hausdorff (1914. It seems therefore more correct to take an intensional view of relations. the couple (x. . 32) gave the definition of the ordered pair (a.11 form of the axiom of reducibility he did not reduce nor otherwise change the propositional-function form *12." . . and must itself be introduced as a primitive idea.. At approximately the same time. b}. were met with skepticism and resistance. x) unless x = y. and to identify them rather with classconcepts than with classes. As noted by Suppes (1960) "This definition . To demonstrate the notion of a relation in extension Russell now embraced the notion of ordered couple: "We may regard a relation .11 (the "two-variable" (relational) version of the axiom of reducibility) by expressing a relation as an ordered pair "using the null set.. Both axioms. But he goes on to say that he would not introduce the ordered couples further into his "symbolic treatment". eliminated axiom *12. . By 1914 Norbert Wiener. {a} }". An attempt to solve the problem of the antinomies led Russell to propose his "doctrine of types" in an appendix B of his 1903 The Principles of Mathematics. But it is necessary to give sense to the couple.asserting that every couple has a relation holding between no other pairs of terms. to distinguish the referent [domain] from the relatum [converse domain]: thus a couple becomes essentially distinct from a class of two terms. like classes. In a few years he would refine this notion and propose in his 1908 The Theory of Types two axioms of reducibility. he proposes his "matrix" and his unpopular axiom of reducibility in their place.. b) as { {a.

substantially as for Frege.e. Schönfinkel's 1924 "provide[s] for . . y'") whose variables x and y must both be arguments and that itself produces an argument (x. and those of a domain of function values . and a set as a function that can take two values. Schönfinkel reduces all mathematics to an extremely compact functional calculus consisting of only three functions: Constancy. fusion (i. Remarkably. They are propositional functions.R 'y =DEF (ιx)(x R y): "the singular object that has a relation R to y". His axiomatization creates two "domains of objects" called "arguments" (I-objects) and "functions" (II-objects). .. the whole sweep of abstract set theory. Russell in his 1920 Introduction to Mathematical Philosophy states that "It should be observed that all mathematical functions result form one-many [sic -contemporary usage is many-one] relations . (read 'the ordered pair x.Schönfinkel's notion of "function" as a many-one "correspondence" 1924 Where exactly the general notion of "function" as a many-one relationship derives from is unclear. It explicitly contains a "contemporary". The crux of the matter is that Schönfinkel lets functions stand as arguments. von Neumann's set theory 1925 By 1925 Abraham Fraenkel (1922) and Thoralf Skolem (1922) had amended Zermelo's set theory of 1908. . propositional and otherwise. the argument domain.(i) the operation [x. A reasonable possibility is the Principia Mathematica notion of "descriptive function" -. however. composition). But von Neumann was not convinced that this axiomatization could not lead to the antinomies.. such that to each argument value there corresponds at most one function value". According to Willard Quine. classes are special sorts of functions. Moses Schonfinkel expressed the notion.. Whatever the case. He introduces two "universal two-variable operations" -. ..)". The latter notion certainly includes the former. All functions. Quine notes that Haskell Curry (1958) carried this work forward "under the head of combinatory logic". To clarify the function pair he notes that "Instead of f(x) we write [f. functions whose values are truth values. . y): ".y)". set-theoretic version of the notion of "function": "[Unlike Zermelo's set theory] [w]e prefer. is to be regarded as a variable in this procedure". and mutual exclusivity. So he proposed his own theory. claiming it to be "well known": "As is well known. . to axiomatize not "set" but "function". (More precisely. . by function we mean in the simplest case a correspondence between the elements of some domain of quantities. y]: ".x] to indicate that f. are for Schönfinkel one-place functions". Functions in this sense are descriptive functions". since a function can be regarded as a set of pairs. where they overlap are the "argument functions" (I-II objects). For Schönfinkel. by 1924. the two notions are completely equivalent. his 1925 An axiomatization of set theory. read 'the value of the function x for the argument y) and (ii) the operation (x.. just like x.

. in Russell's first of all . they denote certain categories of expressions which serve to compose logical and mathematical statements. [these] do not belong to the domain of logic or mathematics. For example.. "primitive formulae". and Halmos (1970).. in Suppes (1960). φ(x) (in Suppes) and S(x) (in Halmos). In their place are "expressions of the object language".. After a discussion of this "functional relation" he asserts that: "The concept of a function which we are considering now differs essentially from the concepts of a sentential [propositional] and of a designatory function .And to avoid the "antinomies of naive set theory. Strictly speaking . by the way.. Kleene 1952 defines the words as follows: "In word languages.. Then a 'predicate' is expressed by an incomplete sentence or sentence skeleton containing an open place. is explained by Alfred Tarski 1946 together with further explanation of the terminology: "An expression such as x is an integer.g. are not very fond of this expression. such as: x + y = 5 are usually referred to by mathematicians as FORMULAE. a proposition is expressed by a sentence. sentential functions and sentences composed entirely of mathematical symbols (and not words of everyday languange). as propositional function". propositional cf his index] FUNCTION.. on replacement of these variables by constants becomes a sentence... While the reader of Suppes (1960) Axiomatic Set Theory or Halmos (1970) Naive Set Theory observes the use of function-symbolism in the axiom of separation. "atomic formulae". e. i. which contains variables and. is called a SENTENTIAL [i.. "___ is a man" expresses a predicate .. define a relation as a set of ordered pairs. For his part Tarski calls the relational form of function a "FUNCTIONAL RELATION or simply a FUNCTION" . In place of "sentential function" we shall sometimes simply say "sentence" --.. The predicate calculus will treat of the logic of predicates in this general sense of 'predicate'. He adopts a notion from Zermelo to restrict these "certain functions" Since 1950 Notion of "function" in contemporary set theory Both axiomatic and naive forms of Zermelo's set theory as modified by Fraenkel (1922) and Skolem (1922) define "function" as a relation.e. .but only in cases where there is no danger of any misunderstanding".e. and "atomic sentences". and define an ordered pair as a set of two "dissymetric" sets. .. because they use the term "function" with a different meaning.. we must forgo treating certain functions as arguments". The predicate is a propositional function of one variable..g. Predicates are often called 'properties' . . But mathematicians. The reason for the disappearance of the words "propositional function" e. . they will see no mention of "proposition" or even "first order predicate calculus".

led to the abstract notion of morphism. the key concept of category theory.. or homomorphisms." Further developments The idea of structure-preserving functions. The term "function" in its new sense. is an expression of a purely logical character. .but they do not denote things treated of in those statements. it designates a certain type of things dealt with in logic and mathematics. on the other hand. More recently.. the concept of functor has been used as an analogue of a function in category theory. .

Because ƒ maps a to 3. with the property that a round trip (a composition) from A to B to A (or from B to A to B) returns each element of the initial set to itself. then inputting y into the inverse function produces the output x.Chapter 2 Inverse Function A function ƒ and its inverse ƒ–1. . and vice versa. Thus. if ƒ is a function from a set A to a set B. then an inverse function for ƒ is a function from B to A. if an input x into the function ƒ produces an output y. In mathematics. the inverse ƒ–1 maps 3 back to a.

Not all functions have an inverse. and the codomain is assumed to be the image. In elementary mathematics. or information-preserving. Every corresponds to at least one . A function with both of these properties is called a bijection. if not otherwise specified. Let ƒ be a function whose domain is the set X. in which case the inverse relation is the inverse function. This is generally stated as two conditions: • • Every corresponds to no more than one . or a surjection. Then. with the property: Stated otherwise. or an injection. each element y ∈ Y must correspond to exactly one element x ∈ X. not to be confused with exponentiation). if it exists. a function is invertible if and only if its inverse relation is a function. . Most functions encountered in elementary calculus do not have an inverse. the inverse of ƒ is the function ƒ−1 with domain Y and codomain X. then ƒ–1 maps Y back to X. so the above is often stated as "a function is bijective if and only if it has an inverse function". the domain is often assumed to be the real numbers. a function ƒ with this property is called one-to-one. Definitions If ƒ maps X to Y. and whose codomain is the set Y. a function ƒ with this property is called onto. the inverse function is then uniquely determined by ƒ and is denoted by ƒ−1 (read f inverse. For this rule to be applicable.A function ƒ that has an inverse is called invertible.

every element of Y must lie in the range of the function ƒ. a function with a codomain is invertible if and only if it is both one-to-one and onto. Y. When using codomains. is called the codomain. Thus. Inverses and composition If ƒ is an invertible function with domain X and range Y. and has the property that every element y ∈ Y corresponds to exactly one element x ∈ X. A function with this property is called onto or a surjection. If the domain is the real numbers. then each element in Y would correspond to two different elements in X (±x). More precisely. If the domain and codomain are both the non-negative numbers. the square of x is not invertible because it is impossible to deduce from its output the sign of its input. and the target. is called the domain of ƒ. Inverses in higher mathematics The definition given above is commonly adopted in calculus. Notice that neither the square root nor the principal square root function is the inverse of x2 because the first is not single-valued. the notation means "ƒ is a function mapping elements of a set X to elements of a set Y". then the function is invertible (by the principal square root) and injective. and the second returns -x when x is negative. and therefore ƒ would not be invertible. Such a function is called non-injective or information-losing.Example: squaring and square root functions The function ƒ(x) = x2 may or may not be invertible. Such a function is called a one-to-one correspondence or a bijection. The source. X. and is considered part of the definition of ƒ. or if the domain is the negative numbers. The codomain contains the range of ƒ as a subset. depending on the domain and codomain. Using the composition of functions we can rewrite this statement as follows: . and it becomes equivalent to the second definition if Y coincides with the codomain of ƒ. then This statement is equivalent to the first of the above-given definitions of the inverse. In higher mathematics. For the inverse to be defined on all of Y. the inverse of a function ƒ: X → Y is required to have domain Y and codomain X.

Note on notation The superscript notation for inverses can sometimes be confused with other uses of superscripts. To avoid confusion. or x + 2. this identity says that the inverse of a function is analogous to a multiplicative inverse. and is called the cosecant. It is usually denoted csc x: . If that is the case then: Then it denotes the inverse function for sin(x) (actually a partial inverse. the notations ƒ[–1] or with the "-1" above the ƒ are sometimes used/needed. then ƒ2(x) = (x + 1) + 1. If we think of composition as a kind of multiplication of functions.where idX is the identity function on the set X. In ƒ−1(x). For example. an inverse trigonometric function is often indicated by the prefix "arc". This explains the origin of the notation ƒ−1. For instance: In trigonometry. the superscript "−1" is not an exponent. for historical reasons. this statement is used as the definition of an inverse morphism. For instance the inverse sine is typically called the arcsine: The function (sin x)–1 is the multiplicative inverse to the sine. especially when dealing with trigonometric and hyperbolic functions. ƒ2 denotes two iterations of the function ƒ. ƒ(n). In category theory. To avoid this confusion. sin2(x) usually does mean the square of sin(x): However. denotes the nth derivative of a function ƒ. if ƒ(x) = x + 1. It is important to realize that ƒ−1(x) is not the same as ƒ(x)−1. A similar notation is used in dynamical systems for iterated functions. with parentheses. the expression sin−1(x) does not always represent the multiplicative inverse to sin(x). In symbols: In calculus. see below).

as in arsinh(x). for the inverse function of sinh(x). Properties Uniqueness If an inverse function exists for a given function ƒ. it is unique: it must be the inverse relation.Hyperbolic functions behave similarly. Symmetry There is a symmetry between a function and its inverse. The property of symmetry can be concisely expressed by the following formula: . then the inverse of ƒ−1 is the original function ƒ. In symbols: This follows because inversion of relations is an involution. This statement is an obvious consequence of the deduction that for ƒ to be invertible it must be injective (first definition of the inverse) or bijective (second definition). using the prefix "ar". Specifically. and csch(x) for the multiplicative inverse of sinh(x). if the inverse of ƒ is ƒ−1.

Such a function is called an involution.Inverse of a composition The inverse of g o ƒ is ƒ–1 o g–1. we must first undo ƒ and then undo g. and let g(x) = 3x. For example. a function ƒ: X → X is equal to its own inverse if and only if the composition ƒ o ƒ is equal to idx. to undo g followed by ƒ. let ƒ(x) = x + 5. and then divide by three: This is the composition (g–1 o ƒ–1) (y). The inverse of a composition of functions is given by the formula Notice that the order of ƒ and g have been reversed. Self-inverses If X is a set. . Then the composition ƒ o g is the function that first multiplies by three and then adds five: To reverse this process. we must first subtract five. then the identity function on X is its own inverse: More generally.

For example. p ≠ 0 y>0 y > 0 and a > 0 various restrictions (see table below) e ax trigonometric functions y1/p (i. such as: A function ƒ from the real numbers to the real numbers possesses an inverse as long as it is one-to-one. y ≥ 0 in general.e. The following table shows several standard functions and their inverses: Function ƒ(x) x+a a–x mx 1/x x2 x3 x p x Inverse ƒ−1(y) y–a a–y y/m 1/y Notes m≠0 x. is to solve the equation y = ƒ(x) for x. if ƒ is the function then we must solve the equation y = (2x + 8)3 for x: . y ≠ 0 x. Such functions are often defined through formulas. if it exists.Inverses in calculus Single-variable calculus is primarily concerned with functions that map real numbers to real numbers. y ≥ 0 only no restriction on x and y x. as long as the graph of the function passes the horizontal line test. ) ln y loga y inverse trigonometric functions Formula for the inverse One approach to finding a formula for ƒ−1. i.e.

For example. if ƒ is the function then ƒ is one-to-one. The dotted line is y = x.Thus the inverse function ƒ−1 is given by the formula Sometimes the inverse of a function cannot be expressed by a formula. If ƒ and ƒ−1 are inverses. There is no simple formula for this inverse. since the equation y = x + sin x cannot be solved algebraically for x. Graph of the inverse The graphs of y = ƒ(x) and y = ƒ–1(x). and therefore possesses an inverse function ƒ−1. then the graph of the function is the same as the graph of the equation .

This is identical to the equation y = ƒ(x) that defines the graph of ƒ. For example. the Jacobian of ƒ−1 at ƒ(p) is the matrix inverse of the Jacobian of ƒ at p. then the formula above can be written This result follows from the chain rule. the function is invertible. then the inverse ƒ−1 will be differentiable as long as ƒ′(x) ≠ 0. Inverses and derivatives A continuous function ƒ is one-to-one (and hence invertible) if and only if it is either strictly increasing or decreasing (with no local maxima or minima). except that the roles of x and y have been reversed. The derivative of the inverse is given by the inverse function theorem: If we set x = ƒ–1(y). This is equivalent to reflecting the graph across the line y = x. let ƒ be the function that converts a temperature in degrees Celsius to a temperature in degrees Fahrenheit: then its inverse function converts degrees Fahrenheit to degrees Celsius: . since the derivative ƒ′(x) = 3x2 + 1 is always positive. In this case. The inverse function theorem can be generalized to functions of several variables. Real-world examples For example. a differentiable function ƒ: Rn → Rn is invertible in a neighborhood of a point p as long as the Jacobian matrix of ƒ at p is invertible. Specifically. If the function ƒ is differentiable. Thus the graph of ƒ−1 can be obtained from the graph of ƒ by switching the positions of the x and y axes.

For example. then we do have an inverse function. suppose ƒ assigns each child in a family its birth year. As well. But if each child was born in a separate year. if a year is given in which no child was born then a child cannot be named. However. An inverse function would output which child was born in a given year. Generalizations Partial inverses The square root of x is a partial inverse to ƒ(x) = x2. For example. the function . Even if a function ƒ is not one-to-one.Or. if the family has twins (or triplets) then the output cannot be known when the input is the common birth year. it may be possible to define a partial inverse of ƒ by restricting the domain. and if we restrict attention to the three years in which a child was born.

there is no need to restrict the domain if we are content with the inverse being a multivalued function: The inverse of this cubic function has three branches. . However. since x2 = (–x)2.g. the inverse of a cubic function with a local maximum and a local minimum has three branches (see the picture above). For example. then the inverse is the negative of the square root of x. For a continuous function on the real line. in which case (If we instead restrict to the domain x ≤ 0. The most important branch of a multivalued function (e. Sometimes this multivalued inverse is called the full inverse of ƒ. and its value at y is called the principal value of ƒ−1(y). the function becomes one-to-one if we restrict to the domain x ≥ 0. and the portions (such as √x and −√x) are called branches.is not one-to-one. one branch is required between each pair of local extrema.) Alternatively. the positive square root) is called the principal branch.

This is considered the principal branch of the inverse sine. so the principal value of the inverse sine is always between –π⁄2 and π⁄2. These considerations are particularly important for defining the inverses of trigonometric functions. For example.The arcsine is a partial inverse of the sine function. and the corresponding partial inverse is called the arcsine. a left inverse for ƒ (or retraction of ƒ) is a function g: Y → X such that . since for every real x (and more generally sin(x + 2πn) = sin(x) for every integer n). However. π⁄2]. the sine is one-to-one on the interval [–π⁄2. the sine function is not one-to-one. The following table describes the principal branch of each inverse trigonometric function: function Range of usual principal value sin−1 –π⁄2 ≤ sin−1(x) ≤ π⁄2 cos−1 0 ≤ cos−1(x) ≤π tan−1 –π⁄2 < tan−1(x) < π⁄2 cot−1 0 < cot−1(x) < π sec−1 0 ≤ sec−1(x) < π⁄2 or π⁄2 < sec−1(x) ≤ π csc−1 −π⁄2 ≤ csc−1(x) < 0 or 0 < csc−1(x) ≤ π⁄2 Left and right inverses If ƒ: X → Y.

A function ƒ has a left inverse if and only if it is injective. if g is a left inverse for ƒ. but may take any values for elements of Y not in the range.∞) denote the squaring map. then g may or may not be a right inverse for ƒ. Then ƒ(g(x))=x for all x in [0. A function ƒ has a right inverse if and only if it is surjective (though constructing such an inverse in general requires the axiom of choice). the preimage (or inverse image) of an element y ∈ Y is the set of all elements of X that map to y: The preimage of y can be thought of as the image of y under the (multivalued) full inverse of the function f. if S is any subset of Y. g(ƒ(-1))=1≠-1.g. h(y) may be any of the elements of x that map to y under ƒ. g is a right inverse to ƒ. that is. then g is not necessarily a left inverse for ƒ. the preimage of S is the set of all elements of X that map to S: .. Preimages If ƒ: X → Y is any function (not necessarily invertible). the function h satisfies the rule Thus. For example let ƒ:R→[0. since.∞). A right inverse for ƒ (or section of ƒ) is a function h: Y → X such that That is. and if g is a right inverse for ƒ. An inverse which is both a left and right inverse must be unique. such that ƒ(x)=x2 for all x in R. However. g is not a left inverse to ƒ. otherwise not. Likewise. e. and let g:[0. the function g satisfies the rule Thus.That is.∞)→R denote the square root map. Similarly. such that g(x)=√x for all x≥0. g must equal the inverse of ƒ on the range of ƒ.

take a function ƒ: R → R. where ƒ: x → x2. When Y is the set of real numbers. Yet preimages may be defined for subsets of the codomain: The preimage of a single element y ∈ Y – a singleton set {y} – is sometimes called the fiber of y. it is common to refer to ƒ−1(y) as a level set. This function is not invertible for reasons discussed above.For example. .

Tables of special functions Many special functions appear as solutions of differential equations or integrals of elementary functions. Because symmetries of differential equations are essential to both physics and mathematics. Not all such systems have efficient algorithms for the evaluation. then arguments. functional analysis. elementary functions are also considered as special functions. open parenthesis. exp. Functions with established international notations are sin. physics. In particular. tables of integrals usually include descriptions of special functions. as well as certain topics in mathematical physics. There is no general formal definition. the standard notation is used for indication of a special function: the name of function. and erfc. the integral representation of special functions. Notations used in special functions In most cases. cos. if any. or other applications. subscripts. and tables of special functions include most important integrals. separated with comma. the theory of special functions is closely related to the theory of Lie groups and Lie algebras. . erf. Such a notation allows easy translation of the expressions to algorithmic languages avoiding ambiguities. but the list of mathematical functions contains functions which are commonly accepted as special. at least. and then close parenthesis. especially in the complex plane. Symbolic computation engines usually recognize the majority of special functions.Chapter 3 Special Functions & Implicit and Explicit Functions Special functions Special functions are particular mathematical functions which have more or less established names and notations due to their importance in mathematical analysis. Therefore.

could be written as handbooks to all the basic identities of the theory. and not . In addition. Subscripts are often used to indicate arguments. History of special functions Classical theory While trigonometry can be codified. In a few cases. the search for a complete and unified theory of special functions has continued since the nineteenth century. usually. but never usually means . refer to the same function. treatises that were essentially complete. The high point of special function theory in the period 1850-1900 was the theory of elliptic functions. the semicolon (. typically integers. tan or tg (especially in Russian literature). the translation to algorithmic languages admits ambiguity and may lead to confusion. although they may behave badly in the case of complex argument(s). arctg.) or even backslash (\) is used as a separator. . Various representations can be used for the evaluation. . or tan − 1. the tangent function may be denoted Tan. the simplest way to evaluate a function is to expand it into a Taylor series. Evaluation of special functions Most special functions are considered as a function of a complex variable.Sometimes. the differential and integral representations are known and the expansion to the Taylor or asymptotic series are available. Examples include: • • • usually indicates is typically . sometimes there exist relations with other special functions. . depending on the context. as it is inconsistent with the others. log or ln. Superscripts may indicate not only exponentiation. as was clear already to expert mathematicians of the eighteenth century (if not before). However. such representation may converge slowly if at all. In algorithmic languages. a special function has several names. The natural logarithm can be called as Log. this one typically causes the most confusion. the singularities and cuts are described. but modification of a function. In this case. a complicated special function can be expressed in terms of simpler functions. They are analytic. arctangent may be called atan. Bessel functions may be written . such as that of Tannery and Molk. For example. rational approximations are typically used. They were based on techniques from complex analysis.

as for the familiar logarithm tables. N. and reduction of as many functions as possible to the given function. The later Bateman manuscript project. In the days before the electronic computer. was a fundamental tool. of extended tables of its values. Macdonald helped to open up large and active new fields with the typical special function flavour. explain what a spherical function can be in general. the special functions were in the particular province of applied mathematics. from 1950 onwards substantial parts of classical theory could be recast in terms of Lie groups. by hand. discovery of infinite series or other analytical expression allowing rapid calculation. Further. The classic Whittaker and Watson textbook sought to unify the theory by using complex variables. Watson tome A Treatise on the Theory of Bessel Functions pushed the techniques as far as possible for one important type that particularly admitted asymptotics to be studied. the ultimate compliment to a special function was the computation. and in particular their representation theory. There is not a real conflict between these approaches. Lie groups. Contemporary theories The modern theory of orthogonal polynomials is of a definite but limited scope. . the G. there are approaches typical of the interests of pure mathematics: asymptotic analysis. Twentieth century The twentieth century saw several waves of interest in special function theory. Conjectures of Ian G. Hypergeometric series became an intricate theory. and the discovery of symmetry principles and other structure behind the façade of endless formulae in rows. The aspects of the theory that then mattered might then be two: • • for numerical analysis. under the editorship of Arthur Erdélyi. In contrast. attempted to be encyclopedic. applications to the physical sciences and engineering determined the relative importance of functions. but it is worth noting other motivations. analytic continuation and monodromy in the complex plane.From that time onwards it would be assumed that analytic function theory. This was a capital-intensive process. and came around the time when electronic computation was coming to the fore and tabulation ceased to be the main issue. in fact. For a long time. which had already unified the trigonometric and exponential functions. work on algebraic combinatorics also revived interest in older parts of the theory. one might say. intended to make the function available by look-up. in need of later conceptual arrangement. Changing and fixed motivations Of course the wish for a broad theory including as many as possible of the known special functions has its intellectual appeal. The end of the century also saw a very detailed discussion of spherical harmonics.

it is still possible to work with implicit functions. Nevertheless. the graph defined by R(x. Special functions in number theory In number theory.y) = 0.y) = 0 will overlap locally with the graph of a function y = f(x). In other situations. Even if it is possible to rearrange this equation to obtain y as an explicit function f(x). in many situations. can be performed with relative ease using implicit differentiation. . By contrast. Implicit functions can often be useful in situations where it is inconvenient to solve explicitly an equation of the form R(x. certain special functions have traditionally been studied. such as differentiation. Geometrically. then one can in principle solve this equation for y. as well as some new ones. To give a function f explicitly is to provide a prescription for determining the output value of the function y in terms of the input value x: y = f(x). Implicit and explicit functions In mathematics. such as particular Dirichlet series and modular forms. at least over some small interval. That is. the function is implicit if the value of y is obtained from x by solving an equation of the form: R(x. an implicit function is a function in which the dependent variable has not been given "explicitly" in terms of the independent variable. Almost all aspects of special function theory are reflected there. the equation R(x. it may not be desirable to do so since the expression of f may be much more complicated than the expression of R. It states that if the equation R(x.Difference equations have begun to take their place besides differential equations as a source for special functions. Some techniques from calculus. it is defined as the level set of a function in two variables: one variable or the other may determine the other. y) = 0 satisfies some mild conditions on its partial derivatives.y) = 0 for y in terms of x. The implicit function theorem provides a link between implicit and explicit functions. but one is not given an explicit formula for one in terms of the other.y) = 0 may fail to define a function at all. and rather defines a kind of multiple-valued function. such as came out of the monstrous moonshine theory.

Many of these iterative methods are based on some form of Newton's method.y) = 0 where R is a polynomial of two variables. Another example is an implicit function given by x − C(y) = 0 where C is a cubic polynomial having a "hump" in its graph.Various numerical methods exist for solving the equation R(x. Solving for y gives Note that there are two "branches" to the implicit function: one where the sign is positive and the other where it is negative. so that a prescribed accuracy can be achieved. then the inverse function of f is a solution of the equation for y in terms of x. the circle equation being one prominent example. Intuitively. Many of these methods are iterative in that they produce successively better approximations. Examples. 1. y) = 0 has a graph that is the graph of a function. an inverse function is obtained from f by interchanging the roles of the dependent and independent variables. 2. the inverse function is the solution y of the equation R(x. The product log is an implicit function given by x − y ey = 0. Caveats Not every equation R(x. Stated another way. Thus. for an . The natural logarithm y = ln(x) is the solution of the equation x − ey = 0.y) = x − f(y) = 0. A simple example of an algebraic function is given by the unit circle: x2 + y2 − 1 = 0. Examples Inverse functions Implicit functions commonly arise as one way of describing the notion of an inverse function.y)=0 to find an approximation to the implicit function y. Algebraic functions play an important role in mathematical analysis and algebraic geometry. Algebraic functions An algebraic function is a solution y for an equation R(x. If f is a function.

However. various constraints are frequently imposed on the allowable sorts of equations or on the domain. Examples 1. y) = 0. When we have an equation R(x. y can be given as a function of x implicitly rather than explicitly. one can differentiate the equation: Solving for : . In order to avoid a problem like this. sometimes it is simpler to differentiate R(x.implicit function to be a true function it might be necessary to use just part of the graph. Implicit differentiation In calculus. Alternatively. the implicit equation x = 0 does not define a function at all. Consider for example This function normally can be manipulated by using algebra to change this equation to an explicit function: Differentiation then gives . As explained in the introduction. An implicit function can sometimes be successfully defined as a true function only after "zooming in" on some part of the x-axis and "cutting away" some unwanted function branches. A resulting formula may only then qualify as a legitimate explicit function. The implicit function theorem provides a uniform way of handling these sorts of pathologies. y) with respect to x and then solve for dy/dx. we may be able to solve it for y and then differentiate. a method called implicit differentiation makes use of the chain rule to differentiate implicitly defined functions. it is a vertical line. The defining equation R = 0 can also have other pathologies. For example.

3. one would have to obtain (via algebra) and then differentiate this function. An example of such a case is the implicit function y5 − y = x.2. It is impossible to express y explicitly as a function of x and dy/dx therefore this cannot be found by explicit differentiation. An example of an implicit function. Using the implicit method. This creates two derivatives: one for y > 0 and another for y < 0. Sometimes standard explicit differentiation cannot be used and. in order to obtain the derivative. dy/dx can be expressed: factoring out shows that which yields the final answer where . for which implicit differentiation might be easier than attempting to use explicit differentiation. is In order to differentiate this explicitly with respect to x. thus. another method such as implicit differentiation must be employed. One might find it substantially easier to implicitly differentiate the implicit function.

then the equation F(x. then M in some small enough neighbourhood of (a. In less technical language.b) is given by a parametrization (x.. .":§ 11. and (a. y) = 0: Marginal rate of substitution In economics. Implicit function theorem It can be shown that if R(x. the implicit derivative (or rather. Fy indicates the derivative of F with respect to x and y The above formula comes from using the generalized chain rule to obtain the total derivative—with respect to x—of both sides of F(x. . −1 times the implicit derivative) is interpreted as the marginal rate of substitution of the two variables: how much more of y one must receive in order to be indifferent to a loss of 1 unit of x. In the standard case where we are given an equation F(x.b) is a point of this submanifold such that the tangent space there is not vertical (that is ).b).y) = 0 the condition on F can be checked by means of partial derivatives.Formula for two variables "The Implicit Function Theorem states that if F is defined on an open disk containing (a. unless the tangent to the supposed graph would be vertical. and Fx and Fy are continuous on the disk..b) = 0.y) is given by a smooth submanifold M in .5 Fx. where F(a. implicit functions exist and can be differentiated.f(x)) where f is a smooth function.b) and the derivative of this function is given by.y) = 0 defines y as a function of x near the point (a. when the level set is an indifference curve.

the functions f: X → Y and g: Y → Z can be composed by computing the output of g when it has an argument of f(x) instead of x. For instance. In mathematics. they may be safely left off. if f. function composition is the application of one function to the results of another. composition of functions will not be commutative. "g after f". That is. or just "g of f". Since there is no distinction between the choices of placement of parentheses. In general.Chapter 4 Function Composition g ∘ f. Commutativity is a special property. where the parentheses serve to indicate that composition is to be performed first for the parenthesized functions. then z is a function of x. . then f ∘ (g ∘ h) = (f ∘ g) ∘ h. or "g composed with f". g. the composition of f and g. Thus one obtains a composite function g ∘ f: X → Z defined by (g ∘ f)(x) = g(f(x)) for all x in X. The functions g and f are said to commute with each other if g ∘ f = f ∘ g. Intuitively. (g ∘ f)(c) = #. "g following f". and h are three functions with suitably chosen domains and codomains. For example. The notation g ∘ f is read as "g circle f". if z is a function g of y and y is a function f of x. The composition of functions is always associative.

attained only by particular functions. (For trigonometric functions. If admits an inverse function. e. By convention. Functional powers If Thus: then may compose with itself. Considering functions as special cases of relations (namely functional relations). this superscript notation represents standard exponentiation when used with trigonometric functions: sin2(x) = sin(x) · sin(x). at least for positive exponents. Then (c ∘ h)(t) describes the oxygen concentration around the plane at time t. Example As an example. But a function always commutes with its inverse to produce the identity mapping. this is sometimes denoted . there is a risk of confusion. For example. Derivatives of compositions involving differentiable functions can be found using the chain rule. the identity map on the domain of . one can analogously define composition of relations. only when . f 2(x) = f(x) · f(x). as f n could also stand for the n-fold product of f. which gives the formula for in terms of and . negative functional powers are defined as the opposite power of the inverse function. suppose that an airplane's elevation at time t is given by the function h(t) and that the oxygen concentration at elevation x is given by the function c(x). For example. and often in special circumstances. Note: If f takes its values in a ring (in particular for real or complex-valued f). The structures given by composition are axiomatized and generalized in category theory. in trigonometry. Repeated composition of a function with itself is called function iteration. . Higher derivatives of such functions are given by Faà di Bruno's formula. usually the latter is meant.g. The functional powers • • for natural follow immediately. .

meaning first do f then do g. an expression for f in g(x) = f r(x) can be derived from the rule for g given non-integer values of r. such as f ∘ f ∘ g ∘ f. potentially complicated chains of these functions composed together. composition monoids can have remarkably complicated structure. They write "xf" for "f(x)" and "(xf)g" for "g(f(x))". In general. This is called fractional iteration. This alternative notation is called postfix notation. For instance. g: X → X having the same domain and codomain. Another example would be that where f is the successor function.. Such long chains have the algebraic structure of a monoid. thus making • . Alternative notations • • Many mathematicians omit the composition symbol.However. writing gf for g ∘ f. some mathematicians decided that writing "g ∘ f" to mean "first apply f. a half iterate of a function f is a function g satisfying g(g(x)) = f(x).g. for instance. f r(x) = x + r. in this case. and one says that the group is generated by these functions. This idea can be generalized so that the iteration count becomes a continuous parameter. One particular notable example is the de Rham curve. Then one can form long. called transformation monoid or composition monoid. then the set of all possible combinations of these functions forms a transformation group. The order is important because matrix multiplication is noncommutative. This can be more natural and seem simpler than writing functions on the left in some areas – in linear algebra. then apply g" was too confusing and decided to change notations. Successive transformations applying and composing to the right agrees with the left-to-right reading sequence. Iterated functions and flows occur naturally in the study of fractals and dynamical systems. In some cases. The set of all bijective functions f: X → X form a group with respect to the composition operator. for negative exponents (especially −1). This is the symmetric group. The set of all functions f: X → X is called the full transformation semigroup on X. tan−1 = arctan (≠ 1/tan). in keeping with the order the symbols occur in postfix notation. If the functions are bijective. Mathematicians who use postfix notation may write "fg". such a system is called a flow. Composition monoids Suppose one has two (or more) functions f: X → X. e. where x is a row vector and f and g denote matrices and the composition is by matrix multiplication. In the mid-20th century. it nevertheless usually refers to the inverse function. also sometimes called the composition group.

Since all functions are binary relations. the composition operator Cg is defined as that operator which maps functions to functions as Composition operators are studied in the field of operator theory. it is correct to use the fat semicolon for function composition as well. To distinguish the left composition operator from a text semicolon.the notation "fg" ambiguous. Composition operator Given a function g. thereby disambiguating the order of composition. . Computer scientists may write "f.g" for this. in the Z notation a fat semicolon ⨟ (U+2A1F) is used for left relation composition.

then M becomes continuous function. if M(t) denotes the amount of money in a bank account at time t.Chapter 5 Continuous Function In mathematics. especially in domain theory. so the function M(t) is discontinuous. In fact. In addition. A continuous function with a continuous inverse function is called "bicontinuous". This function is continuous. for instance the set of points of time at 4:00 PM on business days. . one considers a notion of continuity known as Scott continuity. a continuous function is a function for which. here we discusses the definition for the more general case of functions between two metric spaces.) Real-valued continuous functions Historical infinitesimal definition Cauchy defined continuity of a function in the following intuitive terms: an infinitesimal change in the independent variable corresponds to an infinitesimal change of the dependent variable. The introductory portion here focuses on the special case where the inputs and outputs of functions are real numbers. consider the function h(t) which describes the height of a growing flower at time t. Otherwise. As an example. as every function whose domain is a discrete subset of reals is. if one assumes a discrete set as the domain of function M. intuitively. Continuity of functions is one of the core concepts of topology. (However. By contrast. An intuitive (though imprecise) idea of continuity is given by the common statement that a continuous function is a function whose graph can be drawn without lifting the chalk from the blackboard. In order theory. then the function jumps whenever money is deposited or withdrawn. a function is said to be "discontinuous". which is treated in full generality below. small changes in the input result in small changes in the output. there is a dictum of classical physics which states that in nature everything is continuous.

The function ƒ is said to be continuous at the point c if the following holds: For any number ε > 0. D of R. the graph is a single unbroken curve with no "holes" or "jumps". we say that the function f is continuous at some point c of its domain if. like the functions h and M above.: . and g2 (continuity of curvature). for example. α)(Ω) occurs in the definition of a more subtle concept. Similarly. there exists some number δ > 0 such that for all x in the domain of ƒ with c − δ < x < c + δ. C²(Ω) for the twice-differentiable functions whose second derivative is continuous. then this condition is vacuously true. that of Hölder continuity. every function whose domain is the set of all integers is continuous. Again consider a function ƒ that maps a set of real numbers to another set of real numbers. and suppose c is an element of the domain of ƒ. More generally. these three levels are sometimes called g0 (continuity of position). g1 (continuity of tangency). Such a function can be represented by a graph in the Cartesian plane. in mathematical notation.Definition in terms of limits Suppose we have a function that maps real numbers to real numbers and whose domain is some interval. Weierstrass definition (epsilon-delta) of continuous functions Without resorting to limits. The notation C(Ω) or C0(Ω) is sometimes used to denote the set of all continuous functions with domain Ω. however small. C1(Ω) is used to denote the set of differentiable functions whose derivative is continuous. continuity of ƒ : I → D at c ∈ I means that for every ε > 0 there exists a δ > 0 such that for all x ∈ I. The notation C(n. In the field of computer graphics. one can define continuity of real functions as follows. the value of ƒ(x) satisfies Alternatively written: Given subsets I. roughly speaking. we say that a function is continuous on some subset of its domain if it is continuous at every point of that subset. and only if. Thus. We call a function continuous if and only if it is continuous at every point of its domain. the function is continuous if. and so on. In general. the following holds: • The limit of f(x) as x approaches c through domain of f does exist and is equal to f(c). If the point c in the domain of f is not a limit point of the domain. since x cannot approach c through values not equal c. .

A form of this epsilon-delta definition of continuity was first given by Bernard Bolzano in 1817. this is generalized by the definition of continuity of a function with respect to a basis for the topology. A real function ƒ is continuous if for any sequence (xn) such that it holds that (We assume that all the points xn as well as L belong to the domain of ƒ. and we can do that no matter how small the ƒ(x) neighborhood is. here the metric topology. which is just called continuity. briefly. but in the case of global continuity of real functions it was proved by Wacław Sierpiński that the axiom of choice is not actually needed. we simply need to choose a small enough neighborhood for the x values around c. the concept analogous to Heine definition of continuity is called sequential continuity. More intuitively. it preserves limits. However. and only if. not only the natural numbers). hence representative points in each neighborhood are enough to ensure continuity). Heine definition of continuity The following definition of continuity is due to Heine. The usual (easier) proof makes use of the axiom of choice. .) One can say. then the resulting concept is equivalent to the general notion of continuity in topology. Sequences are sufficient on metric spaces because they are first-countable spaces (every point has a countable neighborhood basis. though the formal definition and the distinction between pointwise continuity and uniform continuity were first given by Karl Weierstrass. ƒ is then continuous at c. we can say that if we want to get all the ƒ(x) values to stay in some small neighborhood around ƒ(c). but general topological spaces are not first-countable. that a function is continuous if. Weierstrass's and Heine's definitions of continuity are equivalent on the reals. Preliminary forms of a related definition of the limit were given by Cauchy. if instead of sequences one uses nets (sets indexed by a directed set. and nets must be used. In general. the condition of sequential continuity is weaker than the analogue of Cauchy continuity. hence sequences do not suffice. In modern terms. In more general setting of topological spaces.

Definition using oscillation The failure of a function to be continuous at a point is quantified by its oscillation. A benefit of this definition is that it quantifies discontinuity: the oscillation gives how much the function is discontinuous at a point. This definition is useful in descriptive set theory to study the set of discontinuities and continuous points – the continuous points are the intersection of the sets where the oscillation is less than ε (hence a Gδ set) – and gives a very quick proof of one direction of the Lebesgue integrability condition. in symbols. Continuity can also be defined in terms of oscillation: a function ƒ is continuous at a point x0 if and only if the oscillation is zero. ωf(x0) = 0. .

then the oscillation is at least ε0. On the other hand. The question of continuity at x= 2 does not arise. In other words. The real line is augmented by the addition of infinite and infinitesimal numbers to form the hyperreal numbers. so each has domain R \ {0} of real numbers except 0.The oscillation is equivalence to the ε-δ definition by a simple re-arrangement. the notion of a continuous function as one whose graph you can draw without taking your pencil off the paper is not quite correct. the resulting function will not be continuous. Another example of a discontinuous function is the signum or sign function. Dirichlet's function . Intuitively we can think of this type of discontinuity as a sudden jump in function values. since x = 0 is neither in the domain of f nor in the domain of g. In nonstandard analysis. If a function has a domain which is not an interval. since no matter what value is assigned at 0. Consider the functions f(x) = 1/x and g(x) = (sin x)/x. The function f cannot be extended to a continuous function whose domain is R. Definition using the hyperreals Non-standard analysis is a way of making Newton-Leibniz-style infinitesimals mathematically rigorous. the oscillation is 0. An example of a discontinuous function is the function f defined by f(x) = 1 if x > 0. A function ƒ from the reals to the reals is continuous if its natural extension to the hyperreals has the property that for real x and infinitesimal dx. continuity can be defined as follows. An example of a rational continuous function is f(x)=1⁄x-2. Pick for instance ε = 1⁄2. are not necessarily continuous on all of R. and by using a limit (lim sup. The question of continuity at x = 0 does not arise. The exponential functions. square root function. A more complicated example of a discontinuous function is Thomae's function. g can be extended continuously to R by defining its value at 0 to be 1. lim inf) to define oscillation: if (at a given point) for a given ε0 there is no δ that satisfies the ε-δ definition. Rational functions. giving a modern expression to AugustinLouis Cauchy's definition of continuity. Examples • • • • • • • • All polynomial functions are continuous. Neither function is defined at x = 0. The oscillation definition can be naturally generalized to maps from a topological space to a metric space. since the limit of g at 0 is 1. however. f(x) = 0 if x ≤ 0. There is no δ-neighborhood around x = 0 that will force all the f(x) values to be within ε of f(0). an infinitesimal increment of the independent variable corresponds to an infinitesimal change of the dependent variable. logarithms. trigonometric functions and absolute value function are continuous. since x = 2 is not in the domain of f. and conversely if for every ε there is a desired δ. ƒ(x+dx) − ƒ(x) is infinitesimal. and each function is continuous.

is nowhere continuous. based on the real number property of completeness. b] and f(a) and f(b) differ in sign. but then any such x does not belong to the domain of the function f/g. Consider for instance the absolute value function at c = 0. at some point c in [a. As a consequence.5 m between the ages of two and six years.e. b]. b] such that f(c) = k. . the child's height must have been 1. then.in other words . b] and k is some number between f(a) and f(b).25 m.b] (or any closed and bounded set) and is continuous there.) The composition f o g of two continuous functions is continuous.b]. Hence f/g is continuous on its entire domain.is continuous. the continuous function f(x) = 1/x. For example. defined on the open interval (0. and states: If the real-valued function f is continuous on the closed interval [a.b) (or any set that is not both closed and bounded). then f + g. fg. or . i. Intermediate value theorem The intermediate value theorem is an existence theorem. in general. then it is also continuous at c. does not attain a maximum. for example. true if the function is defined on an open interval (a.1). as. being unbounded above. if a child grows from 1 m to 1. Extreme value theorem The extreme value theorem states that if a function f is defined on a closed interval [a. f(c) must equal zero.b] with f(c) ≥ f(x) for all x ∈ [a. These statements are not. then. Facts about continuous functions If two functions f and g are continuous. if f is continuous on [a. then there is some number c in [a. (Note. If a function is differentiable at some point c of its domain. The only possible points x of discontinuity of f/g are the solutions of the equation g(x) = 0. The same is true of the minimum of f. and f/g are continuous. there exists c ∈ [a. The converse is not true: a function that is continuous at c need not be differentiable there. at some time between two and six years of age. then the function attains its maximum.

dX) to another metric space (Y. Continuous functions transform limits into limits. f(c)) < ε. the value of ƒ(x) will satisfy Notice that x must be larger than c. Likewise a left-continuous function is a function which is continuous at all points when approached from the left. the formal definition is similar to the definition above for a continuous function but modified as follows: The function ƒ is said to be right-continuous at the point c if the following holds: For any number ε > 0 however small. dY). Then f is continuous at the point c in X if for any positive real number ε. A function is continuous if and only if it is both right-continuous and left-continuous. If x were also allowed to take values less than c. This restriction makes it possible for the function to have a discontinuity at c. there exists a positive real number δ such that all x in X satisfying dX(x. A right-continuous function is a function which is continuous at all points when approached from the right. This can also be formulated in terms of sequences and limits: the function f is continuous at the point c if for every sequence (xn) in X with limit lim xn = c. we have lim f(xn) = f(c). Continuous functions between metric spaces Now consider a function f from one metric space (X.Directional continuity A right continuous function A left continuous function A function may happen to be continuous in only one direction. that is on the right of c. c) < δ will also satisfy dY(f(x). that is. Technically. as pictured. either from the "left" or from the "right". . there exists some number δ > 0 such that for all x in the domain with c < x < c + δ. this would be the definition of continuity. but still be right continuous at c. c − δ < x < c.

If f is continuous at every x ∈ X. However. and c is in the domain of f. . The set of points at which a function between metric spaces is continuous is a Gδ set – this follows from the ε-δ definition of continuity. the intuition is that no matter how "small" V becomes. Although this definition appears complex. then we simply say f is continuous. suppose we have a function f from X to Y. which has a notion of distance. it is equivalent to consider the neighbourhood system of open balls centered at x and f(x) instead of all neighborhoods. Continuous functions between topological spaces Continuity of a function at a point The above definitions of continuous functions can be generalized to functions from one topological space to another in a natural way. In a metric space. We say f is continuous at x for some x ∈ X if for any neighborhood V of f(x). we can always find a U containing x that will map inside it.This latter condition can be weakened as follows: f is continuous at the point c if and only if for every convergent sequence (xn) in X with limit c. is continuous if and only if for every open set V ⊆ Y. Y are topological spaces. Instead. there is a neighborhood U of x such that f(U) ⊆ V. This leads to the standard ε-δ definition of a continuous function from real analysis. Continuous functions transform convergent sequences into Cauchy sequences. the inverse image is open. where X and Y are topological spaces. the sequence (f(xn)) is a Cauchy sequence. which says roughly that a function is continuous if all points close to x map to points close to f(x). this definition is often difficult to use directly. however. where X. This only really makes sense in a metric space. a function f : X → Y.

Neighborhood definition Definitions based on preimages are often difficult to use directly.Note. however. Definitions Several equivalent definitions for a topological structure exist and thus there are several equivalent ways to define a continuous function. which has a notion of distance. then we simply say f is continuous. We say f is continuous at x for some x ∈ X if for any neighborhood V of f(x). there is a neighborhood U of x such that f(U) ⊆ V. In a metric space. which says that preimages (or inverse images) of closed sets are closed. which says roughly that a function is continuous if all points close to x map to points close to f(x). the intuition is that no matter how "small" V becomes. Although this definition appears complicated. . it is still true that f is continuous at a if and only if the limit of f as x approaches a is f(a). Instead. Similar to the open set formulation is the closed set formulation. it is equivalent to consider the neighbourhood system of open balls centered at x and f(x) instead of all neighborhoods. This leads to the standard δ-ε definition of a continuous function from real analysis. If f is continuous at every x ∈ X. every function is continuous. where X and Y are topological spaces. Open and closed set definition The most common notion of continuity in topology defines continuous functions as those functions for which the preimages(or inverse images) of open sets are open. we can always find a U containing x that will map inside it. that if the target space is Hausdorff. however. At an isolated point. This only really makes sense in a metric space. suppose we have a function f : X → Y.

then the converse also holds: any function preserving sequential limits is continuous.int ') where int and int ' are two interior operators then a function is continuous if for all subsets A of X . For non first-countable spaces. known as nets. cl ') where cl and cl ' are two closure operators then a function is continuous if for all subsets A of X One might therefore suspect that given two topological spaces (X. one specifies also when a point is the limit of more general sets of points indexed by a directed set. every function is continuous. Continuous functions preserve limits of nets. a function f : X → Y is sequentially continuous if whenever a sequence (xn) in X converges to a limit x. however. In particular. preservation of limits is also sufficient. Sequences and nets In several contexts. A function is continuous only if it takes limits of sequences to limits of sequences.int) and (X ' .) This motivates the consideration of nets instead of sequences in general topological spaces. Thus sequentially continuous functions "preserve sequential limits". the topology of a space is conveniently specified in terms of limit points. in the latter. In the former case. but for some spaces that are too large in some sense. Every continuous function is sequentially continuous.cl) and (X '. it is still true that f is continuous at a if and only if the limit of f as x approaches a is f(a). if X is a metric space. (The spaces for which the two properties are equivalent are called sequential spaces. If X is a first-countable space. sequential continuity might be strictly weaker than continuity. and preservation of nets is a necessary and sufficient condition. Closure operator definition Given two topological spaces (X.Note. At an isolated point. In detail. the sequence (f(xn)) converges to f(x). sequential continuity and continuity are equivalent. that if the target space is Hausdorff. In many instances. and in fact this property characterizes continuous functions. a function may preserve all limits of sequences yet still fail to be continuous. this is accomplished by specifying when a point is the limit of a sequence.

Useful properties of continuous maps Some facts about continuous maps between topological spaces: • • If f : X → Y and g : Y → Z are continuous. we must resort to inverse images: given two topological spaces (X. Instead.δ') where δ and δ' are two closeness relations then a function is continuous if for all points x and of X and all subsets A of X. neither of these conditions is either necessary or sufficient for continuity. This is another way of writing the closure operator definition. . then so is the composition g ∘ f : X → Z.cl ') where cl and cl ' are two closure operators then a function is continuous if for all subsets A of X ' Closeness relation definition Given two topological spaces (X. then f(X) is compact.int ') where int and int ' are two interior operators then a function is continuous if for all subsets A of X ' We can also write that given two topological spaces (X.int) and (X ' .cl) and (X ' .δ) and (X' . If f : X → Y is continuous and o X is compact.or perhaps if however.

• X is connected. Thus the initial topology can be characterized as the coarsest topology on S which makes f continuous. Thus we can consider the continuity of a given function a topological property. If S has an existing topology. depending only on the topologies of its domain and codomain spaces. if an open map f has an inverse. any function whose range is indiscrete is continuous. then f(X) is Lindelöf. then it is automatically a homeomorphism. X is separable. Thus the final topology can be characterized as the finest topology on S which makes f continuous. for a function f from a set S to a topological space. a partial ordering can be defined on the possible topologies on X. viewed as a subset of X. If the domain set is given the indiscrete topology and the range set is at least T0. that inverse is open. then f(X) is connected. and if a continuous map g has an inverse. X is Lindelöf. this special function is called a homeomorphism. Given a set X. f is continuous with respect to this topology if and only if the existing topology is finer than the initial topology on S. that inverse is continuous. Dually. then f(X) is path-connected. The identity map idX : (X. this topology is canonically identified with the quotient topology under the equivalence relation defined by f. If f is surjective. For a function f from a topological space X to a set S. this topology is canonically identified with the subspace topology of S. If S has an existing topology. Symmetric to the concept of a continuous map is an open map. Conversely. This construction can be generalized to an arbitrary family of functions S → X. If it is. τ1) is continuous if and only if τ1 ⊆ τ2 o o o o Other notes If a set is given the discrete topology. X is path-connected. one defines the initial topology on S by letting the open sets of S be those subsets A of S for which f(A) is open in X. one defines the final topology on S by letting the open sets of S be those subsets A of S for which f−1(A) is open in X. then it has an inverse function. The inverse of a continuous bijection is open. f is continuous with respect to this topology if and only if the existing topology is coarser than the final topology on S. all functions with that space as a domain are continuous. then the only continuous functions are the constant functions. for which images of open sets are open. . If a function is a bijection. In fact. τ2) → (X. but need not be continuous. then f(X) is separable. A continuous function between two topological spaces stays continuous if we strengthen the topology of the domain space or weaken the topology of the codomain space. If f is injective. If a continuous bijection has as its domain a compact space and its codomain is Hausdorff. This construction can be generalized to an arbitrary family of functions X → S.

bk) for all k. b) whenever there are sequences (ak)i and (bk)i in A which converge to a and b respectively for which R(ak. Let X be a complete lattice. Continuity space A continuity space is a generalization of metric spaces and posets. Clearly. . then a function f : X → X is continuous if. continuity of a function between posets is Scott continuity. for each subset Y of X. if one treats R as a characteristic function in two variables. we have sup f(Y) = f(sup Y).Continuous functions between partially ordered sets In order theory. which uses the concept of quantales. this definition of continuous is identical to that for continuous functions. Continuous binary relation A binary relation R on A is continuous if R(a. and that can be used to unify the notions of metric spaces and domains.

For example. The multiplicity of a prime factor p in n. In number theory. Totally additive is also used in this sense by analogy with totally multiplicative functions. but not vice versa. Every completely additive function is additive. this is Cauchy's functional equation. even when they are not co-prime. depending on the specific field of application. the function of the product is the sum of the functions: f(ab) = f(a) + f(b). If f is a completely additive function then f(1) = 0. an additive function is an arithmetic function f(n) of the positive integer n such that whenever a and b are coprime. Completely additive An additive function f(n) is said to be completely additive if f(ab) = f(a) + f(b) holds for all positive integers a and b. that is the largest exponent m for which pm divides n. When the domain is the real numbers. . Examples Example of arithmetic functions which are completely additive are: • • The restriction of the logarithmic function to N.Chapter 6 Additive Function In mathematics the term additive function has two different definitions. In algebra an additive function (or additive map) is a function that preserves the addition operation: f(x + y) = f(x) + f(y) for any two elements x and y in the domain. any linear map is additive.

972.858.279) = 3 ω(54.032.000) = Ω(24 · 53) = Ω(24) + Ω(53) = 4 + 3 = 7 Ω(2.002) = 4 Ω(2.972.the sum of primes dividing n counting multiplicity.000) = ω(24 · 53) = ω(24) + ω(53) = 1 + 1 = 2 ω(2.327. sometimes called the "Big Omega function" (sequence A001222 in OEIS).650.327.802.302) = 1780417 a0(20.415) = 7 Example of arithmetic functions which are additive but not completely additive are: • ω(n).003) = 2003 a0(54.003) = 1 ω(54.972. Ω(1) = 0. For example: ω(4) = 1 ω(20) = ω(22·5) = 2 ω(27) = 1 ω(144) = ω(24 · 32) = ω(24) + ω(32) = 1 + 1 = 2 ω(2. counting multiple factors multiple times.032. since 1 has no prime factors Ω(20) = Ω(2·2·5) = 3 Ω(4) = 2 Ω(27) = 3 Ω(144) = Ω(24 · 32) = Ω(24) + Ω(32) = 4 + 2 = 6 Ω(2.972.279) = 3 Ω(54. sometimes called sopfr(n).650.704.858.279) = 1240658 a0(54.802.• a0(n) .002) = 4 ω(2. defined as the total number of prime factors of n.858.858.858.858.302) = 6 Ω(20.415) = 1240681 • The function Ω(n).972. For example: a0(4) = 2 + 2 = 4 a0(20) = a0(22 · 5) = 2 + 2+ 5 = 9 a0(27) = 3 + 3 + 3 = 9 a0(144) = a0(24 · 32) = a0(24) + a0(32) = 8 + 6 = 14 a0(2.032.704.001) = 3 ω(2.032.302) = 5 .032.001) = 3 Ω(2.003) = 1 Ω(54.032. For example. the potency of n or the integer logarithm of n (sequence A001414 in OEIS). defined as the total number of different prime factors of n (sequence A001221 in OEIS).000) = a0(24 · 53) = a0(24) + a0(53) = 8 + 15 = 23 a0(2.972.

972.858. One such example is g(n) = 2f(n). with the property that whenever a and b are coprime we have: g(ab) = g(a) × g(b).704.327. .327.650.415) = 5 • a1(n) . For example: a1(1) = 0 a1(4) = 2 a1(20) = 2 + 5 = 7 a1(27) = 3 a1(144) = a1(24 · 32) = a1(24) + a1(32) = 2 + 3 = 5 a1(2.the sum of the distinct primes dividing n.002) = 33 a1(2.032.ω(20.032.858.802.003) = 2003 a1(54.302) = 1780410 a1(20.279) = 1238665 a1(54.972.001) = 55 a1(2.802.704.650. sometimes called sopf(n) (sequence A008472 in OEIS).415) = 1238677 Multiplicative functions From any additive function f(n) it is easy to create a related multiplicative function g(n) i.000) = a1(24 · 53) = a1(24) + a1(53) = 2 + 5 = 7 a1(2.e.

An algebraic function in n variables is similarly defined as a function y which solves a polynomial equation in n + 1 variables: It is normally assumed that p should be an irreducible polynomial.Chapter 7 Algebraic Function In mathematics.. Formally. it becomes necessary to introduce ideas relating to Riemann surfaces or more generally algebraic varieties. . Thus an algebraic function is most naturally considered as a multiple valued function. an algebraic function is informally a function that satisfies a polynomial equation whose coefficients are themselves polynomials. Consider for example the equation of a circle: This determines y. In order to understand algebraic functions as functions. except only up to an overall sign: However. and sheaf theory..xn). For example. both branches are thought of as belonging to the "function" determined by the polynomial equation. In more precise terms. The existence of an algebraic function is then guaranteed by the implicit function theorem. at least not in the conventional sense.. an algebraic function in n variables over the field K is an element of the algebraic closure of the field of rational functions K(x1. A function which is not algebraic is called a transcendental function.. an algebraic function may not be a function at all. an algebraic function in one variable x is a solution y for an equation where the coefficients ai(x) are polynomial functions of x.

For supposing that y is a solution of for each value of x. since polynomials are simply the solutions for y of the equation More generally. being the solution of Moreover. note that any polynomial is an algebraic function. y = x2 fails the horizontal line test: it fails to be one-to-one. algebraic functions need not be expressible by radicals. To gain an intuitive understanding. then x is also a solution of this equation for each value of y. Of course. Indeed. . it may be helpful to regard algebraic functions as functions which can be formed by the usual algebraic operations: addition. In this sense. However. the nth root of any polynomial is an algebraic function. solving the equation Surprisingly. because of casus irreducibilis (and more generally the fundamental theorem of Galois theory). not every function has an inverse. Writing x as a function of y gives the inverse function.Algebraic functions in one variable Introduction and overview The informal definition of an algebraic function provides a number of clues about the properties of algebraic functions. also an algebraic function. For example. this is something of an oversimplification. division. the inverse function of an algebraic function is an algebraic function. algebraic functions are often not true functions at all. First. interchanging the roles of x and y and gathering terms. multiplication. The inverse is the algebraic "function" . and taking an nth root. any rational function is algebraic. but instead are multiple valued functions.

x) = 0 is guaranteed to have at least one solution (and in general a number of solutions not exceeding the degree of p in x) for y at each point x. even if one is ultimately interested in real algebraic functions. provided we allow y to assume complex as well as real values. Hence any polynomial relation p(y. A graph of three branches of the algebraic function y. On a more significant theoretical level. For example. Furthermore. the complex numbers are an algebraically closed field. complex numbers enter quite naturally into the study of algebraic functions. problems to do with the domain of an algebraic function can safely be minimized. at least in the multiple-valued sense. Thus. consider the algebraic function determined by the equation Using the cubic formula. there may be no adequate means to express the function in a simple manner without resorting to complex numbers. even though the resulting function is real-valued on the domain of the graph shown. one solution is (the red curve in the accompanying image) There is no way to express this function in terms of real numbers only. . by the fundamental theorem of algebra. First of all. over the domain 3/22/3 < x < 50. the argument principle can be used to show that any algebraic function is in fact an analytic function. In particular. where y3 − xy + 1 = 0.The role of complex numbers From an algebraic perspective. using complex numbers allow one to use the powerful techniques of complex analysis to discuss algebraic functions.

. Hence there are only finitely many such points c1.. . away from the critical points.y) of y has n distinct zeros. Monodromy Note that the foregoing proof of analyticity derived an expression for a system of n different function elements fi(x). (The monodromy action on the universal covering space is related but different notion in the theory of Riemann surfaces. and this occurs only where the highest degree term of p vanishes. this also holds for all x in a neighborhood of x0. In particular.. Choose a system of n nonoverlapping discs Δi containing each of these zeros. We shall show that the algebraic function is analytic in a neighborhood of x0. cm. Suppose that x0 ∈ C is such that the polynomial p(x0.Formally. y) be a complex polynomial in the complex variables x and y. given by the residue theorem: which is an analytic function. and where the discriminant vanishes. Thus the entire function associated to the fi has at worst algebraic poles and ordinary algebraic branchings over the critical points.y) has only one root in Δi. The monodromy group acts by permuting the factors. and thus forms the monodromy representation of the Galois group of p. we have since the fi are by definition the distinct zeros of p. A critical point is a point where the number of distinct zeros is smaller than the degree of p. p(x. A close analysis of the properties of the function elements fi near the critical points can be used to show that the monodromy cover is ramified over the critical points (and possibly the point at infinity). let p(x. provided that x is not a critical point of p(x.) History The ideas surrounding algebraic functions go back at least as far as René Descartes. y). Note that. Then by the argument principle By continuity. The first discussion of algebraic functions appears to have been in Edward Waring's 1794 An Essay on the Principles of Human Knowledge in which he writes: .

be an algebraic function of the abscissa x. . by the common methods of division and extraction of roots. reduce it into an infinite series ascending or descending according to the dimensions of x.let a quantity denoting the ordinate. and then find the integral of each of the resulting terms.

Functions of each type are infinitely differentiable. . The set of all real analytic functions on a given set D is often denoted by Cω(D). categories that are similar in some ways. . a function ƒ is real analytic on an open set D in the real line if for any x0 in D one can write in which the coefficients a0. Definitions Formally.. an analytic function is a function that is locally given by a convergent power series. a1. A function ƒ defined on some subset of the real line is said to be real analytic at a point x if there is a neighborhood D of x on which ƒ is real analytic. an analytic function is an infinitely differentiable function such that the Taylor series at any point x0 in its domain converges to ƒ(x) for x in a neighborhood of x0. Alternatively. but different in others.Chapter 8 Analytic Function In mathematics. but complex analytic functions exhibit properties that do not hold generally for real analytic functions. are real numbers and the series is convergent to ƒ(x) for x in a neighborhood of x0. A function is analytic if and only if it is equal to its Taylor series in some neighborhood of every point.. There exist both real analytic functions and complex analytic functions.

"real" with "complex" and "real line" with "complex plane. Any Taylor series for this function converges not only for x close enough to x0 (as in the definition) but for all values of x (real or complex)." Examples Most special functions are analytic (at least in some range of the complex plane). This is because if a polynomial has degree n. 3) For every compact set there exists a constant C such that for every and every non-negative integer k the following estimate holds: The real analyticity of a function ƒ at a given point x can be characterized using the FBI transform. The exponential function is analytic. and so this series will be trivially convergent.The definition of a complex analytic function is obtained by replacing. and the power functions are analytic on any open set of their domain. every polynomial is its own Maclaurin series. • • Typical examples of functions that are not analytic are: • The absolute value function when defined on the set of real numbers or complex numbers is not everywhere analytic because it is not differentiable at 0. 2) There is a complex analytic extension of ƒ to an open set which contains D. Typical examples of analytic functions are: • Any polynomial (real or complex) is an analytic function. Piecewise defined functions (functions given by different formulas in different regions) are typically not analytic where the pieces meet. • Alternate characterizations If ƒ is an infinitely differentiable function defined on an open set following conditions are equivalent. The complex conjugate function is not complex analytic. Furthermore. although its restriction to the real line is the identity function and therefore real analytic. in the definitions above. . The trigonometric functions. logarithm. then the 1) ƒ is real analytic. any terms of degree larger than n in its Taylor series expansion will vanish. .

then ƒ is zero everywhere on the connected component containing the accumulation point. The situation is quite different when one considers complex analytic functions and complex derivatives. These statements imply that while analytic functions do have more degrees of freedom than polynomials. Also. the analytic functions are sparse compared to all infinitely differentiable functions. infinitely differentiable. compare complex derivatives below. then ƒ is identically zero on the connected component of D containing r. and the space of real analytic functions is a proper subspace of the space of smooth functions. The reciprocal of an analytic function that is nowhere zero is analytic. Properties of analytic functions • • • • The sums. The converse is not true. and compositions of analytic functions are analytic. and are thus much more easily characterized. A similar but weaker statement holds for analytic functions.Complex analytic functions are exactly equivalent to holomorphic functions. they are still quite rigid. in fact. as is the inverse of an invertible analytic function whose derivative is nowhere zero. Consequently. in a certain sense. if all the derivatives of an analytic function at a point are zero. A polynomial cannot be zero at too many points unless it is the zero polynomial (more precisely. the term analytic function is synonymous with holomorphic function. the set A(Ω) of all analytic functions u : Ω → C is a Fréchet space with respect to the uniform convergence on compact sets. In other words. If the set of zeros of an analytic function ƒ has an accumulation point inside its domain. The set of all bounded analytic functions with the supremum norm is a Banach space. in complex analysis. It can be proved that any complex function differentiable (in the complex sense) in an open set is analytic. that is. if (rn) is a sequence of distinct numbers such that ƒ(rn) = 0 for all n and this sequence converges to a point r in the domain of D. . or C∞). products. In fact there are many such functions. (Note that this differentiability is in the sense of real variables. any analytic function (real or complex) is infinitely differentiable (also known as smooth. For any open set Ω ⊆ C. Any analytic function is smooth. the function is constant on the corresponding connected component.) There exist smooth real functions which are not analytic. the number of zeros is at most the degree of the polynomial). Analyticity and differentiability As noted above. The fact that uniform limits on compact sets of analytic functions are analytic is an easy consequence of Morera's theorem.

This statement for real analytic functions (with open ball meaning an open interval of the real line rather than an open disk of the complex plane) is not true in general. Analytic functions of several variables have some of the same properties as analytic functions of one variable. not every real analytic function defined on the whole real line can be extended to a complex function defined on the whole complex plane. . its power series expansion at x0 is convergent in the whole ball. as it is not defined for x = ±i. as it has more restrictive necessary conditions and complex analytic functions have more structure than their real-line counterparts. is clearly false.Real versus complex analytic functions Real and complex analytic functions have important differences (one could notice that even from their different relationship with differentiability). For instance. any bounded complex analytic function defined on the whole complex plane is constant. The corresponding statement for real analytic functions. The function ƒ (x) defined in the paragraph above is a counterexample. new and interesting phenomena show up when working in 2 or more dimensions. Any real analytic function on some open set on the real line can be extended to a complex analytic function on some open set of the complex plane. However. zero sets of complex analytic functions in more than one variable are never discrete. especially for complex analytic functions. Analyticity of complex functions is a more restrictive property. since the power series 1 − x2 + x4 − x6. diverges for |x| > 1. this is illustrated by Also.. if a complex analytic function is defined in an open ball around a point x0. with the complex plane replaced by the real line. Analytic functions of several variables One can define analytic functions in several variables by means of power series in those variables.. the function of the example above gives an example for x0 = 0 and a ball of radius exceeding 1. However. According to Liouville's theorem.

Chapter 9

Completely Multiplicative Function and Concave Function

Completely multiplicative function
In number theory, functions of positive integers which respect products are important and are called completely multiplicative functions or totally multiplicative functions. Especially in number theory, a weaker condition is also important, respecting only products of coprime numbers, and such functions are called multiplicative functions. Outside of number theory, the term "multiplicative function" is often taken to be synonymous with "completely multiplicative function" as defined here.

A completely multiplicative function (or totally multiplicative function) is an arithmetic function (that is, a function whose domain is the natural numbers), such that f(1) = 1 and f(ab) = f(a) f(b) holds for all positive integers a and b. Without the requirement that f(1) = 1, one could still have f(1) = 0, but then f(a) = 0 for all positive integers a, so this is not a very strong restriction.

The easiest example of a multiplicative function is a monomial: For any particular positive integer n, define f(a) = an.

A completely multiplicative function is completely determined by its values at the prime numbers, a consequence of the fundamental theorem of arithmetic. Thus, if n is a product of powers of distinct primes, say n = pa qb ..., then f(n) = f(p)a f(q)b ...

Concave function
In mathematics, a concave function is the negative of a convex function. A concave function is also synonymously called concave downwards, concave down, convex cap or upper convex.

A real-valued function f defined on an interval (or on any convex set C of some vector space) is called concave if, for any two points x and y in its domain C and any t in [0,1], we have

A function is called strictly concave if

for any t in (0,1) and x ≠ y. For a function f:R→R, this definition merely states that for every z between x and y, the point (z, f(z) ) on the graph of f is above the straight line joining the points (x, f(x) ) and (y, f(y) )

A function f(x) is a quasiconcave if the upper contour sets of the function are convex sets.

A function f(x) is concave over a convex set if and only if the function −f(x) is a convex function over the set. A differentiable function f is concave on an interval if its derivative function f ′ is monotonically decreasing on that interval: a concave function has a decreasing slope. ("Decreasing" here means "non-increasing", rather than "strictly decreasing", and thus allows zero slopes.) For a twice-differentiable function f, if the second derivative, f ′′(x), is positive (or, if the acceleration is positive), then the graph is convex; if f ′′(x) is negative, then the graph is concave. Points where concavity changes are inflection points. If a convex (i.e., concave upward) function has a "bottom", any point at the bottom is a minimal extremum. If a concave (i.e., concave downward) function has an "apex", any point at the apex is a maximal extremum. If f(x) is twice-differentiable, then f(x) is concave if and only if f ′′(x) is non-positive. If its second derivative is negative then it is strictly concave, but the opposite is not true, as shown by f(x) = -x4. If f is concave and differentiable then

A continuous function on C is concave if and only if for any x and y in C

If a function f is concave, and f(0) ≥ 0, then f is subadditive. Proof:

since f is concave, let y = 0,

• • •

The functions f(x) = − x2 and are concave, as the second derivative is always negative. Any linear function f(x) = ax + b is both concave and convex. The function f(x) = sin(x) is concave on the interval .

• • The function log | B | . . where | B | is the determinant of matrix nonnegativedefinite matrix B. is concave. Practical application: rays bending in Computation of radiowave attenuation in the atmosphere.

.Chapter 10 Convex Function Convex function on an interval A function (in black) is convex if and only if the region above its graph (in green) is a convex set.

y) is the slope of the red line in the above drawing. . Pictorially. note also that the function R is symmetric in x. concave upwards.In mathematics. for y fixed (or viceversa). a real-valued function f(x) defined on an interval (or on any convex subset of some vector space) is called convex. then f may fail to be continuous at the endpoints of C (an example is shown in the examples' section). i. . These two definitions are equivalent. one holds if and only if the other one is true. Properties Suppose f is a function of one real variable defined on an interval. for any two points in the interval.y). A function f is said to be (strictly) concave if −f is (strictly) convex. A convex function f defined on some open interval C is continuous on C and Lipschitz continuous on any closed subinterval. f admits left and right derivatives. f is convex if and only if R(x. and these are monotonically non-decreasing. This characterization of convexity is quite useful to prove the following results. otherwise the point may not lie in the function domain.. if for any two points x1 and x2 in its domain X and any A function is called strictly convex if for every . As a consequence.y) is monotonically nondecreasing in x. a function is called 'convex' if the function lies below or on the straight line segment connecting two points. Note that the function must be defined over a convex set. cup. and . Sometimes an alternative definition is used: A function is convex if its epigraph (the set of points lying on or above the graph) is a convex set.e. f is differentiable at all but at most countably many points. If C is closed. and let (note that R(x. concave up or convex .

this gives a practical test for convexity. A strictly convex function will have at most one global minimum. a function whose sublevel sets are convex sets may fail to be a convex function. For example. the second derivative of f(x) = x4 is f "(x) = 12 x2. For example. and f(0) ≤ 0. In particular. This condition is only slightly weaker than convexity.) If a function f is convex. but the converse does not hold. In particular. Any local minimum of a convex function is also a global minimum. a real valued Lebesgue measurable function that is midpoint convex will be convex. the sublevel sets {x | f(x) < a} and {x | f(x) ≤ a} with a ∈ R are convex sets. A twice differentiable function of one variable is convex on an interval if and only if its second derivative is non-negative there. then c is a global minimum of f(x). Jensen's inequality applies to every convex function f. More generally. a continuous. if f '(c) = 0. If a function is differentiable and convex then it is also continuously differentiable. A differentiable function of one variable is convex on an interval if and only if its derivative is monotonically non-decreasing on that interval. A continuously differentiable function of one variable is convex on an interval if and only if the function lies above all of its tangents: for all x and y in the interval. For a convex function f. which is zero for x = 0. a continuous function that is midpoint convex will be convex.A function is midpoint convex on an interval C if for all x and y in C. If its second derivative is positive then it is strictly convex. then f is superadditive on the positive half-axis. A function whose sublevel sets are convex is called a quasiconvex function. Proof: . but x4 is strictly convex. If X is a random variable taking values in the domain of f. However. twice differentiable function of several variables is convex on a convex set if and only if its Hessian matrix is positive semidefinite on the interior of the convex set. then (Here denotes the mathematical expectation.

A third definition for a strongly convex function. for every • Convex function calculus • • • • If f and g are convex functions. then its perspective function g(x. If f(x) is convex.y) is convex in x then is convex in x. Strongly convex functions The concept of strong convexity extends and parametrizes the notion of strict convexity. with parameter m. A differentiable function f is called strongly convex with parameter m > 0 if the following equation holds for all points x. If f is concave and g is convex and non-increasing. is that. then h(x) = g(f(x)) is convex.t) = tf(x / t) (whose domain is ) is convex.g(x)} and h(x) = f(x) + g(x). Convexity is invariant under affine maps: that is.y in its domain: This is equivalent to the following It is not necessary for a function to be differentiable in order to be strongly convex.y in the domain and . provided • • for some x. then h(x) = g(f(x)) is convex. If f and g are convex functions and g is non-decreasing. for all x. A strongly convex function is also strictly convex. but not vice-versa.• since f is convex. if f(x) is convex with . . then so is g(y) = f(Ay + b) with . then so are m(x) = max{f(x). where If f(x. let y = 0.

and strongly convex can be subtle at first glimpse. consider a function f that is strictly convex. where I is the identity and is the Hessian matrix. If the function f is twice continuously differentiable. but not strongly convex. then is just the . and is identical to the definition of a convex function when m = 0. so the condition becomes . Despite this. and suppose there is a sequence of points (xn) such that strongly convex because . which implies the function is convex. . Like strictly convex functions. The distinction between convex. the function is not Strongly convex functions are in general easier to work with than convex or strictly convex functions. since they are a smaller class. Assuming still that the function is twice continuously differentiable. and perhaps strictly convex. If m = 0.Notice that this definition approaches the definition for strict convexity as . strictly convex. it means that ). strongly convex functions have unique minima. Then by the assumption about the eigenvalues. and hence we recover the second strong convexity equation above. If f is twice continuously differentiable and the domain is the real line. then f is strongly convex with parameter m if and only if for all x in the domain. and the inequality means that is positive definite. functions exist that are strictly convex but are not strongly convex for any m > 0 (see example below). then this second derivative means the Hessian is positive semidefinite (or if the domain is the real line. then we can characterize it as follows: convex if and only if strictly convex if for all for all (note: this is sufficient. but not necessary) for all strongly convex if and only if For example. Start by using Taylor's for some (unknown) . This is equivalent to requiring that the minimum eigenvalue of be at least m for all x. Even though will become arbitrarily small. . If the domain is just the real line. we show that the lower bound of Theorem: implies that it is strongly convex.

1] defined by f(0) = f(1) = 1. then f(a + b) = f(a) + f(b). the function g(x) = ef(x) is logarithmically convex if f is a convex function. The absolute value function f(x) = | x | is convex. Every affine function taking values in . It is concave on the interval (-∞. The function f(x) = | x | p for 1 ≤ p is convex. f(x) = 0 for 0 < x < 1 is convex. since if f is linear. It is also strictly convex. The function f(x) = 1/x2. thus it is convex on the set where x ≥ 0 and concave on the set where x ≤ 0. because of the singularity at x = 0. with strong convexity constant 2. The function f with domain [0. +∞). The function f(x) = 1/x has which is greater than 0 if x > 0. +∞) and convex on the interval (-∞. is convex on the interval (0. each function of the form f(x) = aTx + b. Every norm is a convex function. but it is not strongly convex since the second derivative can be arbitrarily close to zero. • • • • • • • • • • • • • . so f(x) is convex on the interval (0. so f is a convex function. The function f(x) = x4 has .0). even though it does not have a derivative at the point x = 0. but not continuous at 0 and 1. +∞). Every linear transformation taking values in is convex but not strictly convex. since f''(x) = ex > 0. The exponential function f(x) = ex is convex.e. so f is a convex function. 1). It is also strongly convex (and hence strictly convex too). Examples of functions that are monotonically increasing but not convex include and g(x) = log(x). It is strictly convex. Examples of functions that are convex but not monotonically increasing include h(x) = x2 and k(x) = − x.. even though the second derivative is not strictly positive at all points. It is not strongly convex. is simultaneously convex and concave. but not convex on the interval (-∞.0). i. The function x3 has second derivative 6x. It is not strictly convex. it is continuous on the open interval (0. This statement also holds if we replace "convex" by "concave". by the triangle inequality and positive homogeneity. with f(0) = +∞.Examples • The function f(x) = x2 has f''(x) = 2 > 0 at all points. More generally.

Chapter 11 Differentiable Function A differentiable function .

and therefore cannot have a break. More generally. As a result. a differentiable function is a function whose derivative exists at each point in its domain. or cusp at this point. The graph of a differentiable function must have a non-vertical tangent line at each point in its domain. This means that the graph of ƒ has a non-vertical tangent line at the point (x0. bend. or any points with a vertical tangent. ƒ(x0)). if x0 is a point in the domain of a function ƒ. the graph of a differentiable function must be relatively smooth. or cusps. . and cannot contain any breaks. In calculus (a branch of mathematics).The absolute value function is not differentiable at x = 0. then ƒ is said to be differentiable at x0 if the derivative ƒ′(x0) is defined. bends.

it is possible for the derivative to have an essential discontinuity. this means that differentiable functions are very atypical among continuous functions. The converse does not hold: a continuous function need not be differentiable. However. then ƒ must also be continuous at x0. If ƒ is differentiable at a point x0. or vertical tangent may be continuous. In particular. but fails to be differentiable at the location of the anomaly. and is itself a continuous function. Most functions which occur in practice have derivatives at all points or at almost every point. a result of Stefan Banach states that the set of functions which have a derivative at some point is a meager set in the space of all continuous functions. Informally. but is not differentiable at any point. the function . For example. a function with a bend.Differentiability and continuity The Weierstrass function is continuous. cusp. Though the derivative of a differentiable function never has a jump discontinuity. The first known example of a function that is continuous everywhere but differentiable nowhere is the Weierstrass function. any differentiable function must be continuous at every point in its domain. For example. Differentiability classes A function ƒ is said to be continuously differentiable if the derivative ƒ′(x) exists.

. but all of the partial derivatives and directional derivatives exist at this point. then the function must be differentiable at that point. and is in fact of class C1. For a continuous example.. ƒ(k)(x) all exist and are continuous.. but again all of the partial derivatives and directional derivatives exist.is differentiable at 0 (with the derivative being 0). It is known that if the partial derivatives of a function all exist and are continuous in a neighborhood of a point. then all of the partial derivatives must exist at x0. the function ƒ: R2 → R defined by is not differentiable at (0. More generally. Note that existence of the partial derivatives (or even all of the directional derivatives) does not guarantee that a function is differentiable at a point. but the derivative is not continuous at this point. . in which case the linear map J is given by the Jacobian matrix. 0). For example. the function is not differentiable at (0. A function is of class C2 if the first and second derivative of the function both exist and are continuous. Sometimes continuously differentiable functions are said to be of class C1. 0). ƒ″(x). a function is said to be of class Ck if the first k derivatives ƒ′(x).. Differentiability in higher dimensions A function f: Rm → Rn is said to be differentiable at a point x0 if there exists a linear map J: Rm → Rn such that If a function is differentiable at x0.

Differentiable functions on manifolds If M is a differentiable manifold.Differentiability in complex analysis In complex analysis. Such a function is necessarily infinitely differentiable. More generally. a real or complex-valued function ƒ on M is said to be differentiable at a point p if it is differentiable with respect to some (or any) coordinate chart defined around p. if M and N are differentiable manifolds. and in fact analytic. a function ƒ: M → N is said to be differentiable at a point p if it is differentiable with respect to some (or any) coordinate charts defined around p and ƒ(p). any function that is complex-differentiable in a neighborhood of a point is called holomorphic. .

arccos(x) is an elementary function. one variable. An example of a function that is not elementary is the error function . This last function is equal to the inverse cosine trigonometric function arccos(x) in the entire complex domain. The roots of equations are the functions implicitly defined as solving a polynomial equation with constant coefficients. For polynomials of degree four and smaller there are explicit formulae for the roots (the formulae are elementary functions).Chapter 12 Elementary Function and Entire Function Elementary function In mathematics. and nth roots through composition and combinations using the four elementary operations (+ – × ÷). logarithms. trigonometric functions and their inverses become included in the elementary functions. Elementary functions were introduced by Joseph Liouville in a series of papers from 1833 to 1841. An algebraic treatment of elementary functions was started by Joseph Fels Ritt in the 1930s. constants. Hence. an elementary function is a function built from a finite number of exponentials. By allowing these functions (and constants) to be complex numbers. Examples Examples of elementary functions include: and .

) The derivation captures the properties of differentiation. two special types of transcendental extensions (the logarithm and the exponential) can be added to the field building a tower containing elementary functions. or is a logarithm. If the base field is over the rationals. (Here ∂u is a new function. the derivation is linear and satisfies the Leibniz product rule An element h is a constant if ∂h = 0. that is. is considered in the context of differential algebra. ∂u = u ∂a for a ∈ F. care must be taken when extending the field to add the needed transcendental constants. A function u of a differential extension F[u] of a differential field F is an elementary function over F if the function u • • • is algebraic over F. (this is Liouville's theorem). or is an exponential. or a function in elementary form. . so that for any two elements of the base field. Differential algebra The mathematical definition of an elementary function. By starting with the field of rational functions. ∂u = ∂a / u for a ∈ F. Sometimes the notation u′ is used.a fact that cannot be seen directly from the definition of elementary function but can be proven using the Risch algorithm. A differential algebra is an algebra with the extra operation of derivation (algebraic version of differentiation). that is. A differential field F is a field F0 (rational functions over the rationals Q for example) together with a derivation map u → ∂u. Using the derivation operation new equations can be written and their solutions used in extensions of the algebra.

except possibly one. A transcendental entire function is an entire function that is not a polynomial. As a consequence of Liouville's theorem. Specifically. including the error function and the trigonometric functions sine and cosine and their hyperbolic counterparts the hyperbolic sine and hyperbolic cosine functions. of degree at least n. with n a natural number and M and R positive constants. Conversely. Thus any nonconstant entire function must have a singularity at the complex point at infinity. for any transcendental entire function f and any complex w there is a sequence (zm)m∈N with . Picard's little theorem is a much stronger result: any non-constant entire function takes on every complex number as value. Typical examples of entire functions are the polynomials and the exponential function. Liouville's theorem may be used to elegantly prove the fundamental theorem of algebra. Liouville's theorem states that any bounded entire function must be constant. with n a natural number and M and R positive constants. The latter exception is illustrated by the exponential function. of degree at most n. any function which is entire on the whole Riemann sphere (complex plane and the point at infinity) is constant. an entire function. and any sums. which never takes on the value 0. any entire function f satisfying the inequality for all z with . is a complexvalued function that is holomorphic over the whole complex plane. The Weierstrass factorization theorem asserts that any entire function can be represented by a product involving its zeroes.Entire function In complex analysis. products and compositions of these. by the Casorati–Weierstrass theorem. is necessarily a polynomial. Neither the natural logarithm nor the square root functions can be continued analytically to an entire function. is necessarily a polynomial. Liouville's theorem is a special case of the following statement: any entire function f satisfying the inequality for all z with . also called an integral function. The entire functions on the complex plane form an integral domain (in fact a Prüfer domain). and . either a pole for a polynomial or an essential singularity for a transcendental entire function. Properties Every entire function can be represented as a power series which converges uniformly on compact sets.

Order and growth
The order (at infinity) of an entire function f(z) is defined using the limit superior as:

where Br is the disk of radius r and one can also define the type:

denotes the supremum norm of f(z) on Br. If

In other words, the order of f(z) is the infimum of all m such that as . The order need not be finite.

Entire functions may grow as fast as any increasing function: for any increasing function there exists an entire function f(z) such that f(x) > g( | x | ) for all real x. Such a function f may be easily found of the form:

, for a conveniently chosen strictly increasing sequence of positive integers nk. Any such sequence defines an entire series f(z); and if it is conveniently chosen, the inequality f(x) > g( | x | ) also holds, for all real x.

Other examples
J. E. Littlewood chose the Weierstrass sigma function as a 'typical' entire function in one of his books. Other examples include the Fresnel integrals, the Jacobi theta function, and the reciprocal Gamma function. The exponential function and the error function are special cases of the Mittag-Leffler function.

Chapter 13

Even and Odd Functions

In mathematics, even functions and odd functions are functions which satisfy particular symmetry relations, with respect to taking additive inverses. They are important in many areas of mathematical analysis, especially the theory of power series and Fourier series. They are named for the parity of the powers of the power functions which satisfy each condition: the function f(x) = xn is an even function if n is an even integer, and it is an odd function if n is an odd integer.

Even functions

ƒ(x) = x2 is an example of an even function. Let f(x) be a real-valued function of a real variable. Then f is even if the following equation holds for all x in the domain of f:

Geometrically, the graph of an even function is symmetric with respect to the y-axis, meaning that its graph remains unchanged after reflection about the y-axis. Examples of even functions are |x|, x2, x4, cos(x), and cosh(x).

Odd functions

ƒ(x) = x3 is an example of an odd function. Again, let f(x) be a real-valued function of a real variable. Then f is odd if the following equation holds for all x in the domain of f:

or Geometrically. meaning that its graph remains unchanged after rotation of 180 degrees about the origin. and erf(x). sin(x). x3. sinh(x). Examples of odd functions are x. Some facts ƒ(x) = x3 + 1 is neither even nor odd . the graph of an odd function has rotational symmetry with respect to the origin.

A function's being odd or even does not imply differentiability. In fact. derivatives and so on may only be used when they can be assumed to exist. The integral of an odd function from −A to +A is zero (where A is finite. The derivative of an even function is odd. The derivative of an odd function is even. The quotient of two odd functions is an even function. The sum of two odd functions is odd. any linear combination of odd functions is odd. The sum of an even and odd function is neither even nor odd. and the function has no vertical asymptotes between −A and A). The Fourier series of a periodic even function includes only cosine terms. The quotient of two even functions is an even function. The product of two odd functions is an even function. the vector space of all real-valued functions is the direct sum of the subspaces of even . Series • • • • The Maclaurin series of an even function includes only even powers. Taylor series. and the odd functions also form a vector space over the reals. The Fourier series of a periodic odd function includes only sine terms. The product of two even functions is an even function. or even continuity. The product of an even function and an odd function is an odd function. and any constant multiple of an odd function is odd. Algebraic structure • Any linear combination of even functions is even. The Maclaurin series of an odd function includes only odd powers. The sum of two even functions is even. and the even functions form a vector space over the reals. The composition of two even functions is even. The composition of any function with an even function is even (but not vice versa). The quotient of an even function and an odd function is an odd function.e. unless one of the functions is identically zero. and the function has no vertical asymptotes between −A and A). For example. Basic properties • • • • • • • • • • • • • • • • • The only function which is both even and odd is the constant function which is identically zero (i. and the composition of two odd functions is odd. the Dirichlet function is even. but is nowhere continuous. and any constant multiple of an even function is even. Similarly.. Properties involving Fourier series. f(x) = 0 for all x). The composition of an even function and an odd function is even. The integral of an even function from −A to +A is twice the integral from 0 to +A (where A is finite.

every function f(x) can be written uniquely as the sum of an even function and an odd function: where is even and is odd. • The even functions form a commutative algebra over the reals.and odd functions. if f is exp. the resulting signal will consist of only odd harmonics of the input sine wave. harmonic distortion occurs when a sine wave signal is sent through a memoryless nonlinear system. o The fundamental is also an odd harmonic. When it is asymmetric. The type of harmonics produced depend on the response function f: • • • When the response function is even. . a system whose output at time t only depends on the input at time t and does not depend on the input at any previous times. that is. In other words. the resulting signal will consist of only even harmonics of the input sine wave. For example. the odd functions do not form an algebra over the reals. the resulting signal may contain either even or odd harmonics. However. o Simple examples are a half-wave rectifier. When it is odd. then fe is cosh and fo is sinh. Such a system is described by a response function Vout(t) = f(Vin(t)). Harmonics In signal processing. and clipping in an asymmetrical class A amplifier. o A simple example is clipping in a symmetric push-pull amplifier. o The output signal will be half-wave symmetric. so will not be present. o A simple example is a full-wave rectifier.

This is usually written as . mathematical physics and the theory of stochastic processes. a harmonic function is a twice continuously differentiable function f : U → R (where U is an open subset of Rn) which satisfies Laplace's equation.e.Chapter 14 Harmonic Function A harmonic function defined on an annulus. In mathematics. i. everywhere on U.

and/or has a constant added. and so the corresponding harmonic function will be proportional to the electrostatic potential due to these charge distributions. Function Singularity Unit point charge at origin x-directed dipole at origin Line of unit charge density on entire z-axis Line of unit charge density on negative z-axis Line of x-directed dipoles on entire z axis Line of x-directed dipoles on negative z axis . The singular points of the harmonic functions below are expressed as "charges" and "charge densities" using the terminology of electrostatics. Each function below will yield another harmonic function when multiplied by a constant. the electric potential between the plates of a capacitor. and the gravity potential of a slab) The function on for n > 2. rotated. linear and affine functions on all of (for example. Examples of harmonic functions of three variables are given in the table below with r2 = x2 + y2 + z2. Also. and the gravity potential due to a long cylindrical mass) • The function Examples of harmonic functions of n variables are: • • The constant. the sum of any two harmonic functions will yield another harmonic function.Examples Examples of harmonic functions of two variables are: • • The real and imaginary part of any holomorphic function The function defined on (e. the electric potential due to a line charge.g. Harmonic functions are determined by their singularities. The inversion of each function will yield another harmonic function which has singularities which are the images of the original singularities in a spherical "mirror".

they can be locally expressed as power series. This sequence is harmonic and converges uniformly to the zero function. a theorem of removal of singularities as well as a Liouville theorem one holds for them in analogy to the corresponding theorems in complex functions theory. of which the Laplacian is a major example. Consider the sequence on ( . The Laplace operator Δ and the partial derivative operator will commute on this class of functions. They are (real) analytic. any harmonic function u on an open set is locally the real part of a holomorphic function. The uniform limit of a convergent sequence of harmonic functions is still harmonic. This is true because any continuous function satisfying the mean value property is harmonic. then all partial derivatives of f are also harmonic functions on U. Conversely. . writing z = x + iy. still harmonic functions in n variables enjoy a number of properties typical of holomorphic functions. they have a maximum principle and a mean-value principle. i. however note that the partial derivatives are not uniformly convergent to the zero function (the derivative of the zero function). the complex function g(z): = ux − iuy is holomorphic in Ω. the harmonic functions are real analogues to holomorphic functions. Properties of harmonic functions Some important properties of harmonic functions can be deduced from Laplace's equation. This is a general fact about elliptic operators. This is immediately seen observing that. 0)× R defined by . Although the above correspondence with holomorphic functions only holds for functions of two real variables.e. differences and scalar multiples of harmonic functions are again harmonic.Remarks The set of harmonic functions on a given open set U can be seen as the kernel of the Laplace operator Δ and is therefore a vector space over R: sums. Therefore. All harmonic functions are analytic. In several ways. This example shows the importance of relying on the mean value property and continuity to argue that the limit is harmonic. because it satisfies the Cauchy-Riemann equations. Connections with complex function theory The real and imaginary part of any holomorphic function yield harmonic functions on R2 (these are said to be a pair of harmonic conjugate functions). If f is a harmonic function on U. and u is the real part of f up to a constant. g has locally a primitive f. as ux is the real part of .

The mean value theorem follows by verifying that the spherical mean of u is constant: which in turn follows by making a change of variable and then applying Green's theorem.Regularity theorem for harmonic functions Harmonic functions are infinitely differentiable. real analytic. In particular. Similar arguments also show that harmonic distributions are.r) is a ball with center x and radius r which is completely contained in the open set . restricted to K. (smooth) harmonic functions (Weyl's lemma). As a consequence of the mean value theorem. harmonic functions are real analytic. other than the exceptional case where f is constant. this average value is also equal to the average value of u in the interior of the ball.ε) and ∫η = 1. . the convolution η∗u is also smooth. then f. attains its maximum and minimum on the boundary of K. in fact. In other words where ωn is the volume of the unit ball in n dimensions and σ is the n−1 dimensional surface measure. if η is an integrable radial function supported in B(0. then the value u(x) of a harmonic function at the center of the ball is given by the average value of u on the surface of the ball. Mean value property If B(x. by the Poisson integral representation). In fact. by taking η to be a C∞ function. If U is connected. then provided that B(x. u is preserved by the convolution of a harmonic function u with any radial function η with total integral one.ε) ⊂ Ω. this means that f cannot have local maxima or minima. Similar properties can be shown for subharmonic functions. and therefore harmonic functions are smooth throughout their domains (in fact. More precisely. Maximum principle Harmonic functions satisfy the following maximum principle: if K is any compact subset of U.

For locally integrable functions. Harnack's inequality Let u be a non-negative harmonic function in a bounded domain Ω. it follows since the mean value property implies that u is unchanged when convolved with any radial mollifier of total integral one. that is then f extends to a harmonic function on Ω (compare Riemann's theorem for functions of a complex variable). This follows for C2 functions again by the method of spherical means. then f is constant (compare Liouville's theorem for functions of a complex variable). which is less singular at x0 than the fundamental solution. If f is a harmonic function defined on a dotted open subset of Rn.The converse to the mean value theorem also holds: all locally integrable functions satisfying the (volume) mean-value property are infinitely differentiable and harmonic functions as well. but convolutions with mollifiers are smooth and so the C2 result can still be applied. Generalizations Weakly harmonic function A function (or. Liouville's theorem If f is a harmonic function defined on all of Rn which is bounded above or bounded below. more generally. a distribution) is weakly harmonic if it satisfies Laplace's equation . Then for every connected set Harnack's inequality holds for some constant C that depends only on V and Ω. Removal of singularities The following principle of removal of singularities holds for harmonic functions.

With the exception of the mean value theorem. although other properties of harmonic functions may fail. for all Harmonic functions on manifolds Harmonic functions can be defined on an arbitrary Riemannian manifold. equivalently. and the Harnack inequality.in a weak sense (or. and so also is smooth. A weakly harmonic function coincides almost everywhere with a strongly harmonic function. More generally. in the interior of any ball in its domain. it is possible to define harmonic vector-valued functions. . and is in particular smooth. which are critical points of a generalized Dirichlet energy functional (this includes harmonic functions as a special case. using the Laplace-Beltrami operator Δ. This is Weyl's lemma. and it is related to the study of cohomology. in the sense of distributions). these are easy consequences of the corresponding results for general linear elliptic partial differential equations of the second order. representing harmonic functions in the Sobolev space H1(Ω) as the minimizers of the Dirichlet energy integral with respect to local variations. including the mean value theorem (over geodesic balls). Subharmonic functions A C2 function that satisfies is called subharmonic. its graph lies below that of the harmonic function interpolating its boundary values on the ball. One of which is Dirichlet's principle. or harmonic maps of two Riemannian manifolds. that is. all functions holds for all such that or equivalently. Also. This condition guarantees that the maximum principle will hold. the maximum principle. a function is subharmonic if and only if. There are other weak formulations of Laplace's equation that are often useful. Harmonic forms One generalization of the study of harmonic functions is the study of harmonic forms on Riemannian manifolds. A weakly harmonic distribution is precisely the distribution associated to a strongly harmonic function. a result known as Dirichlet principle). In this context. a function is called harmonic if Many of the properties of harmonic functions on domains in Euclidean space carry over to this more general setting.

. that is. a map from an interval in R to a Riemannian manifold. which are precisely the harmonic immersions of a surface into three-dimensional Euclidean space. More generally. minimal submanifolds are harmonic immersions of one manifold in another. For example. is a harmonic map if and only if it is a geodesic. Important special cases of harmonic maps between manifolds include minimal surfaces. then a harmonic map u : M → N is defined to be a stationary point of the Dirichlet energy in which du : TM → TN is the differential of u.These kind of harmonic maps appear in the theory of minimal surfaces. Harmonic coordinates are a harmonic diffeomorphism from a manifold to an open subset of a Euclidean space of the same dimension. and the norm is that induced by the metric on M and that on N on the tensor product bundle T∗M⊗u−1TN. a curve. Harmonic maps between manifolds If M and N are two Riemannian manifolds.

for it implies that any . The existence of a complex derivative is a very strong condition. holomorphic functions are the central objects of study in complex analysis. A holomorphic function is a complex-valued function of one or more complex variables that is complex-differentiable in a neighborhood of every point in its domain. In mathematics.Chapter 15 Holomorphic Function A rectangular grid (top) and its image under a holomorphic function f (bottom).

This concept of complex differentiability shares several properties with real differentiability: it is linear and obeys the product rule. the derivative of ƒ at a point z0 in its domain is defined by the limit This is the same as the definition of the derivative for real functions. y) is holomorphic. the limit is taken as the complex number z approaches z0. or of more general type) that is equal to its Taylor series in a neighborhood of each point in its domain. except that all of the quantities are complex. then u and v have first partial derivatives with respect to x and y. we say that ƒ is holomorphic on U. If ƒ is complex differentiable at every point z0 in U. y) + i v(x. The relationship between real differentiability and complex differentiability is the following. The fact that the class of complex analytic functions coincides with the class of holomorphic functions is a major theorem in complex analysis. and must have the same value for any sequence of complex values for z that approach z0 on the complex plane. but differentiable everywhere within some neighbourhood of z0 in the complex plane. we say that ƒ is differentiable at the point z0. The phrase "holomorphic at a point z0" means not just differentiable at z0. complex. If the limit exists. A holomorphic function whose domain is the whole complex plane is called an entire function. If a complex function ƒ(x + i y) = u(x. Definition Given a complex-valued function ƒ of a single complex variable. We say that ƒ is holomorphic on some non-open set A if it is holomorphic in an open set containing A. and chain rule. The term analytic function is often used interchangeably with “holomorphic function”. Holomorphic functions are also sometimes referred to as regular functions or as conformal maps. In particular. although the word “analytic” is also used in a broader sense to describe any function (real. and satisfy the Cauchy– Riemann equations: .holomorphic function is actually infinitely differentiable and equal to its own Taylor series. We say that ƒ is holomorphic at the point z0 if it is holomorphic on some neighborhood of z0. quotient rule.

If one identifies C with R2. and derives from the Greek ὅλος (holos) meaning "entire". Briot (1817–1882) and Bouquet (1819–1895). the term "holomorphic function" is sometimes preferred to "analytic function". and the quotient of two holomorphic functions is holomorphic wherever the denominator is not zero. Today. as the latter is a more general concept. This is also because an important result in complex analysis is that every holomorphic function is complex analytic. The derivative f'(a) can be written as a contour integral using Cauchy's differentiation formula: for any simple loop positively winding once around a. In other words. Properties Because complex differentiation is linear and obeys the product. Every holomorphic function can be separated into its real and imaginary parts. and chain rules. then the holomorphic functions coincide with those functions of two real variables with continuous first derivatives which solve the Cauchy-Riemann equations. and they satisfy the Cauchy–Riemann equations. and for infinitesimal positive loops γ around a. and each of these is a solution of Laplace's equation on R2. the converse is not necessarily true. The term "analytic" is however also in wide use. . then ƒ is holomorphic. y) both u and v are harmonic functions. and μορφή (morphē) meaning "form" or "appearance". y) + i v(x. u and v have first partial derivatives. a fact that does not follow directly from the definitions. if we express a holomorphic function f(z) as u(x. quotient.If continuity is not a given. A simple converse is that if u and v have continuous first partial derivatives and satisfy the Cauchy–Riemann equations. the sums. Terminology The word "holomorphic" was introduced by two of Cauchy's students. is the Looman–Menchoff theorem: if ƒ is continuous. which is much harder to prove. products and compositions of holomorphic functions are holomorphic. A more satisfying converse. a set of two partial differential equations. then ƒ is holomorphic.

From an algebraic point of view. It follows from that df′ is also proportional to dz.) Examples All polynomial functions in z with complex coefficients are holomorphic on C. Fγ(z) is independent of the particular choice of path γ. That is. the fact that d(f dz) = f′ dz ∧ dz = 0 implies that any function f that is holomorphic on the simply connected region U is also integrable on U. a function f is holomorphic at z0 if and only if its exterior derivative df in a neighborhood U of z0 is equal to f′(z) dz for some continuous function f′. Similarly. f coincides with its Taylor series at a in any disk centered at that point and lying within the domain of the function. In fact. Every holomorphic function is analytic. . and so are sine. in light of the Jordan curve theorem and the generalized Stokes' theorem. implying that the derivative f′ is itself holomorphic and thus that f is infinitely differentiable. The function 1/z is holomorphic on {z : z ≠ 0}. the set of holomorphic functions on an open set is a commutative ring and a complex vector space. define . cosine and the exponential function. In fact. and it coincides with its own Taylor series at a in a neighborhood of a. From a geometrical perspective. a holomorphic function f has derivatives of every order at each point a in its domain. (For a path γ from z0 to z lying entirely in U. and thus F(z) is a well-defined function on U having F(z0) = F0 and dF = f dz. The principal branch of the complex logarithm function is holomorphic on the set C \ {z ∈ R : z ≤ 0}. Cauchy's integral formula states that every function holomorphic inside a disk is completely determined by its values on the disk's boundary. holomorphic functions are conformal in the sense that they preserve angles and the shape (but not size) of small figures. with the seminorms being the suprema on compact subsets.In regions where the first derivative is not zero. The square root function can be defined as and is therefore holomorphic wherever the logarithm log(z) is. (The trigonometric functions are in fact closely related to and can be defined via the exponential function using Euler's formula). it is a locally convex topological vector space.

the real part of z and the imaginary part of z are not holomorphic. the Fréchet or Gâteaux derivative can be used to define a notion of a holomorphic function on a Banach space over the field of complex numbers. the absolute value of z. centered at that point) as a convergent power series in the variables. . This condition is stronger than the Cauchy–Riemann equations. For instance. the argument of z. Another typical example of a continuous function which is not holomorphic is complex conjugation. Therefore. in fact it can be stated as follows: A function of several complex variables is holomorphic if and only if it satisfies the Cauchy–Riemann equations and is locally square-integrable. Several variables A complex analytic function of several complex variables is defined to be analytic and holomorphic at a point if it is locally expandable (within a polydisk. a real-valued holomorphic function must be constant. Extension to functional analysis The concept of a holomorphic function can be extended to the infinite-dimensional spaces of functional analysis. a Cartesian product of disks.As a consequence of the Cauchy–Riemann equations.

a slightly more general form of homogeneity is often used.Chapter 16 Homogeneous Function In mathematics. More generally. an homogeneous function is a function with multiplicative scaling behaviour: if the argument is multiplied by a factor. then an homogeneous function from S to W can still be defined by (1). Homogeneous functions can also be defined for vector spaces with the origin deleted. When the vector spaces involved are over the real numbers. . then ƒ is said to be homogeneous of degree k ∈ F if (1) for all nonzero α ∈ F and v ∈ V. requiring only that (1) hold for all α > 0. More precisely. if S ⊂ V is any subset that is invariant under scalar multiplication by elements of the field (a "cone"). a fact that is used in the definition of sheaves on projective space in algebraic geometry. then the result is multiplied by some power of this factor. if ƒ : V → W is a function between two vector spaces over a field F.

v2 ∈ V2.Examples A homogeneous function is not necessarily continuous.e. i. Homogeneous polynomials Monomials in n variables define homogeneous functions ƒ : Fn → F.. This function is homogeneous of order 1. vn ∈ Vn.y) = x if xy > 0 or f(x. as shown by this example..x. It follows that the n-th differential of a function ƒ : X → Y between two Banach spaces X and Y is homogeneous of degree n. since by the definition of linearity for all α ∈ F and v ∈ V.y) = 0 if . any multilinear function ƒ : V1 × V2 × . It is discontinuous at y = 0. Vn → W is homogeneous of degree n. For example. . since by the definition of multilinearity for all α ∈ F and v1 ∈ V1... f(αx. . Linear functions Any linear function ƒ : V → W is homogeneous of degree 1. This is the function f defined by f(x.αy) = αf(x.y) for any real numbers α.. Similarly.y.

. in this example. one of an homogeneous polynomial from a multilinear form and the other of a multilinear form from an homogeneous polynomial. Thus. In finite dimensions. A homogeneous polynomial is a polynomial made up of a sum of monomials of the same degree. is a homogeneous polynomial of degree 5. V → F on the n-th Cartesian product of V. then f/g is homogeneous of degree m − n away from the zeros of g. For example. Polarization A multilinear function g : V × V × . The degree is the sum of the exponents on the variables. they establish an isomorphism of graded vector spaces from the symmetric algebra of V∗ to the algebra of homogeneous polynomials on V. V → F from the n-th Cartesian product of V with itself to the groundfield F gives rise to an homogeneous function ƒ : V → F by evaluating on the diagonal: The resulting function ƒ is a polynomial on the vector space V. the polarization of ƒ is a multilinear function g : V × V × . then given an homogeneous polynomial ƒ of degree n on V. Homogeneous polynomials also define homogeneous functions. Conversely.. . if F has characteristic zero. are mutually inverse to one another.is homogeneous of degree 10 since . 10=5+2+3. if f is homogeneous of degree m and g is homogeneous of degree n... The polarization is defined by These two constructions. Rational functions Rational functions formed as the ratio of two homogeneous polynomials are homogeneous functions off of the affine cone cut out by the zero locus of the denominator.

The natural logarithm f(x) = lnx scales additively and so is not homogeneous. This can be proved by noting that f(5x) = ln5x = ln5 + f(x), f(10x) = ln10 + f(x), and f(15x) = ln15 + f(x). Therefore such that .

Affine functions
The function f(x) = x + 5 does not scale multiplicatively.

Positive homogeneity
In the special case of vector spaces over the real numbers, the notation of positive homogeneity often plays a more important role than homogeneity in the above sense. A function ƒ : V \ {0} → R is positive homogeneous of degree k if

for all α > 0. Here k can be any complex number. A (nonzero) continuous function homogeneous of degree k on Rn \ {0} extends continuously to Rn if and only if Re{k} > 0. Positive homogeneous functions are characterized by Euler's homogeneous function theorem. Suppose that the function ƒ : Rn \ {0} → R is continuously differentiable. Then ƒ is positive homogeneous of degree k if and only if

This result follows at once by differentiating both sides of the equation ƒ(αy) = αkƒ(y) with respect to α and applying the chain rule. The converse holds by integrating. As a consequence, suppose that ƒ : Rn → R is differentiable and homogeneous of degree k. Then its first-order partial derivatives are homogeneous of degree k − 1. The result follows from Euler's theorem by commuting the operator with the partial derivative.

Homogeneous distributions
A compactly supported continuous function ƒ on Rn is homogeneous of degree k if and only if

for all compactly supported test functions φ and nonzero real t. Equivalently, making a change of variable y = tx, ƒ is homogeneous of degree k if and only if

for all t and all test functions φ. The last display makes it possible to define homogeneity of distributions. A distribution S is homogeneous of degree k if

for all nonzero real t and all test functions φ. Here the angle brackets denote the pairing between distributions and test functions, and μt : Rn → Rn is the mapping of scalar multiplication by the real number t.

Application to differential equations
The substitution v = y/x converts the ordinary differential equation

where I and J are homogeneous functions of the same degree, into the separable differential equation

Chapter 17

Indicator Function

The graph of the indicator function of a two-dimensional subset of a square. In mathematics, an indicator function or a characteristic function is a function defined on a set X that indicates membership of an element in a subset A of X, having the value 1 for all elements of A and the value 0 for all elements of X not in A.

The indicator function of a subset A of a set X is a function

defined as

then . also called a bound variable). 1·1 = 1. This mapping is surjective only when A is a proper subset of X. if then .The Iverson bracket allows the equivalent notation. By a similar argument. maps elements of X to the range {0. Basic properties The indicator or characteristic function of a subset A of some set X. . . respectively. to be used instead of The indicator function of A is sometimes denoted or or even . may signify the characteristic function in convex analysis. then In the following. If A and B are two subsets of X. the dot represents multiplication. while mathematicians in other fields are more likely to use the term characteristic function to describe the function which indicates membership in a set. If . 1·0 = 0 etc. The term "characteristic function" has an unrelated meaning in probability theory. probabilists use the term indicator function for the function defined here almost exclusively. (The Greek letter χ because it is the initial letter of the Greek etymon of the word characteristic. A related concept in statistics is that of a dummy variable (this must not be confused with "dummy variables" as that term is usually used in mathematics. "+" and "−" represent addition and subtraction.1}. " " and " " is intersection and union. For this reason.) Remark on notation and terminology • • The notation The notation may signify the identity function.

the indicator random variable otherwise . becomes a random variable whose expected value is equal to the probability of A: then This identity is used in a simple proof of Markov's inequality. variance and covariance Given a probability space is defined by (mean) with if . is clearly a product of 0s and 1s. as a generalization of the inverse of the indicator function in elementary number theory. such as order theory. In many cases. for instance in probability theory: if X is a probability space with probability measure and A is a measurable set.and the "complement" of the indicator function of A i. This is commonly called the generalized Möbius function. As suggested by the previous example. This is one form of the principle of inclusion-exclusion. That is Expanding the product on the left hand side. suppose is a collection of subsets of X. the inverse of the indicator function may be defined. the indicator function is a useful notational device in combinatorics. The notation is used in other places as well.e. where | F | is the cardinality of F. AC is: More generally. Mean. This product has the value 1 at precisely those which belong to none of the sets Ak and is 0 otherwise. For any . the Möbius function.

xn) and φ(x1. etc. . Characteristic function in fuzzy set theory In classical mathematics. in some algebra or structure (usually required to be at least a poset or lattice). "warm". 279ff) mu operators (Kleene (1952)) and the CASE function (p. . . . 1].. AND. xn)=1 if ~R(x1. What appears to the modern reader as the representing function's logical-inversion." (p. . Fuzzy sets model the gradual change in the membership degree seen in many real-world predicates like "tall".. . 229). . . . xn) = 0 if R(x1.(variance) (covariance) Characteristic function in recursion theory. . plays a useful role in Kleene's definition of the logical functions OR. takes on values 0 if the predicate is true and 1 if the predicate is false. For example. . . The Undecidable): "There shall correspond to each class or relation R a representing function φ(x1. 228) and unbounded(p. because the product of characteristic functions φ1*φ2* . Gödel's and Kleene's representing function Kurt Gödel described the representing function in his 1934 paper "On Undecidable Propositions of Formal Mathematical Systems". . characteristic functions of sets only take values 1 (members) or 0 (non-members). (The paper appears on pp. . the representing function is 0 when the function R is "true" or satisfied".e. and IMPLY (p. Such generalized characteristic functions are more usually called membership functions. . and the corresponding "sets" are called fuzzy sets.e. . i. *φn = 0 whenever any one of the functions equals 0. 41-74 in Martin Davis ed. 42. OR φn=0 THEN their product is 0. 228). 227) offers up the same definition in the context of the primitive recursive functions as a function φ of a predicate P. characteristic functions are generalized to take value in the real unit interval [0.. or more generally. xn).(p. the bounded.. "NOT") Stephen Kleene (1952) (p. it plays the role of logical OR: IF φ1=0 OR φ2=0 OR . . In fuzzy set theory. the "~" indicates logical inversion i.

Chapter 18 Injective Function An injective function (not a bijection) Another injective function (is a bijection) .

if a ≠ b. An injective function is called an injection. In particular the identity function X → X is always injective (and in fact bijective). then f(a) ≠ f(b).e. and is also said to be a one-to-one function (not to be confused with one-to-one correspondence.. it may be denoted YX using a notation derived from that used for falling factorial powers. every element of the function's codomain is mapped to by at most one element of its domain. then the function is said to be bijective. using an arrow with a barbed tail. but is used to stress the opposition with multi-valued functions. the number of injections X ↣ Y is yx. i.A non-injective function (this one happens to be a surjection) In mathematics. The function f is injective if for all a and b in A. f(a) = f(b) implies a = b. an injective function from X to Y is denoted f: X ↣ Y. Alternately.e. i. this terminology is also sometimes used to mean "single-valued". this is the case for any function. Definition Let f be a function whose domain is a set A. since if X and Y are finite sets with respectively x and y elements. then a = b. an injective function is a function that preserves distinctness: it never maps distinct elements of its domain to the same element of its codomain. A function f that is not injective is sometimes called many-to-one. a bijective function). If in addition all of the elements in the codomain are in fact mapped to by some element of the domain. In other words. each argument is mapped to at most one value. Examples • For any set X and any subset S of X the inclusion map S → X (which sends any element s of S to itself) is injective. . which are not true functions. Occasionally.) A monomorphism is a generalization of an injective function in category theory. that is. if f(a) = f(b). (However. Equivalently.

f ∘ g. if there is a function g : Y → X such that.Y ∘ g. one can at least obtain a "quasi-inverse" of it. where inclJ. Injections may be made invertible In fact. given f : X → Y. Injections are "reversible" but not always invertible. then g is bijective. More generally. Indeed. since. a function that can be undone or "reversed". to turn an injective function f : X → Y into a bijective (hence invertible) function. Note that g may not be a complete inverse of f because the composition in the other order.• • • • • The function f : R → R defined by f(x) = 2x + 1 is injective. then an injective function f : R → R is one whose graph is never intersected by any horizontal line more than once. In this case. g(0) = g(1). for example. ∞) → R defined by x ↦ ln x is injective. f is called a section of g and g is called a retraction of f. The exponential function exp : R → R defined by exp(x) = ex is injective (but not surjective as no value maps to a negative number). is not necessarily invertible (bijective). This principle is referred to as the horizontal line test. that is a multiple-valued function. because (for example) g(1) = 1 = g(−1). when X and Y are both the real line R. The function g : R → R defined by g(x) = x2 is not injective. Injections can be undone Functions with left inverses are always injections. such as f. In other words.+∞). That is.Y is the inclusion function from J into Y. The function g : R → R defined by g(x) = xn − x is not injective. every injection f with non-empty domain has a left inverse g (in conventional mathematics). let g : X → J such that g(x) = f(x) for all x in X. The natural logarithm function ln : (0. then g is injective. . then f ∘ g is injective. That is. for every x ∈ X g(f(x)) = x (f can be undone by g) then f is injective. Other properties • If f and g are both injective. it suffices to replace its codomain Y by its actual range J = f(X). Although it is impossible to reverse a non-injective (and therefore information-losing) function. if g is redefined so that its domain is the non-negative real numbers [0. may not be the identity on Y. f can be factored as inclJ. However. Conversely.

An injective function which is a homomorphism between two algebraic structures is an embedding. h : W → X. This decomposition is unique up to isomorphism. f : X → Y is injective if and only if. then Y has at least as many elements as X. . then f(A ∩ B) = f(A) ∩ f(B). whenever f ∘ g = f ∘ h. then X and Y has the same cardinal number. then f −1(f(A)) = A. in addition. if. and f may be thought of as the inclusion function of the range h(W) of h as a subset of the codomain Y of h.The composition of two injective functions is injective. then f : X → Y is injective if and only if f is surjective (in which case they are bijective). then g = h. In other words. in the sense of cardinal numbers. then f is injective (but g need not be). Every function h : W → Y can be decomposed as h = f ∘ g for a suitable injection f and surjection g. Thus. (This is known as the Cantor–Bernstein–Schroeder theorem. • • • • • • • • If g ∘ f is injective. A can be recovered from its image f(A). given any functions g. If f : X → Y is injective and A and B are both subsets of X.) If both X and Y are finite with the same number of elements. there is an injection from Y to X. injective functions are precisely the monomorphisms in the category Set of sets. In particular. If f : X → Y is an injective function. If f : X → Y is injective and A is a subset of X.

as such. In probability theory. A function . a function between measurable spaces is said to be measurable if the preimage of each measurable set is measurable. as special care must be taken regarding the σ-algebras involved. This definition can be deceptively simple. If the values of the function lie in an infinite-dimensional vector space instead of R or C. In particular. In contrast.Chapter 19 Measurable Function In mathematics. Specifically. A complex-valued measurable function is defined analogously. functions that are not Lebesgue measurable are generally considered pathological. In practice. such as weak measurability and Bochner measurability. the composition of Lebesgue-measurable functions need not be Lebesgue-measurable. the domain and range represent different σ-algebras on the same underlying set (here is the sigma algebra of Lebesgue measurable sets.Τ) be measurable spaces. some authors use measurable functions to refer only to real-valued measurable functions with respect to the Borel algebra.Σ) and (Y. For instance. the sigma algebra often represents the set of available information. when a function is said to be Lebesgue measurable what is actually meant is that is a measurable function—that is. As a result. and is the Borel algebra on ). Formal definition Let (X. at least in the field of analysis. By convention a topological space is assumed to be equipped with the Borel algebra generated by its open subsets unless otherwise specified. a real-valued measurable function is a function for which the preimage of each Borel set is measurable. they form a natural context for the theory of integration. measurable functions are structurepreserving functions between measurable spaces. and a function (in this context a random variable) is measurable if and only if it represents an outcome that is knowable based on the available information. particularly in measure theory. meaning that X and Y are sets equipped with respective sigma algebras Σ and Τ. Most commonly this space will be the real or complex numbers. analogous to the situation of continuous functions between topological spaces. however. usually other definitions of measurability are used.

a measurable function is nearly a continuous function. and limit inferior of a sequence (viz. • • The (pointwise) supremum. The notion of measurability depends on the sigma algebras Σ and Τ.Σ) and (Y. However.. such as uniform convergence. so long as there is no division by zero. • Random variables are by definition measurable functions defined on sample spaces. and measurable functions are of interest in mathematical analysis because they can be integrated. The composition of measurable functions is measurable. countably many) of real-valued measurable functions are all measurable as well.is said to be measurable if for every . So is the quotient. a measurable function is also called a Borel function. • A Lebesgue measurable function is a measurable function . If a Borel function happens to be a section of some map . Continuous functions are Borel functions but not all Borel functions are continuous.e. where is the sigma algebra of Lebesgue measurable is the Borel algebra on the complex numbers . then so is . it is called a Borel section. Lebesgue sets. But see the caveat regarding Lebesgue-measurable functions in the introduction. if is a measurable function. To emphasize this dependency. limit superior. Properties of measurable functions • The sum and product of two complex-valued measurable functions are measurable.Τ) are Borel spaces. The pointwise limit of a sequence of measurable functions is measurable. note that the corresponding statement for continuous functions requires stronger conditions than pointwise convergence. • . infimum. if and are measurable functions. i. we will write Special measurable functions • If (X..

then the indicator function is non-measurable (where is equipped with the Borel algebra as usual). • So long as there are non-measurable sets in a measure space. since the preimage of any point in the range is some proper. however. and therefore does not lie in Σ. Here 1A is given by • Any non-constant function can be made non-measurable by equipping the domain and range with appropriate σ-algebras. nonempty subset of X.X}. If is an arbitrary nonconstant.Σ) is some measurable space and is a non-measurable set. i. . if . since the preimage of the measurable set {1} is the nonmeasurable set A. there are nonmeasurable functions from that space. it is not difficult to find non-measurable functions. real-valued function.Non-measurable functions Real-valued functions encountered in applications tend to be measurable.e. If (X. then f is non-measurable if X is equipped with the indiscrete algebra Σ = {0.

You're Reading a Free Preview

/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->