Vector Spaces

1. Vector Spaces
2. Subspaces of Vector Spaces
3. Spanning Sets and Linear Independence
4. Basis and Dimension
 Vector Spaces
• Vector spaces:
Let V be a set (vectors) and F be a set of scalers on which two
operations (vector addition and scalar multiplication) are defined.
If the following axioms are satisfied for every u, v, and w in V
and every scalars c and d, then V is called a vector space over F.
Addition:
(1) u+v is in V (closed under addition)
(2) u+v=v+u
(3) u+(v+w)=(u+v)+w
(4) V has a zero vector 0 such that for every u in V, u+0=u
(5) For every u in V, there is a vector in V denoted by –u
such that u+(–u)=0
2/44
Scalar multiplication:
(6) cu is in V (closed under multiplication by a scalar).
(7) c(u  v )  cu  cv

(8) (c  d )u  cu  du

(9) c( du)  (cd )u

(10) 1(u)  u

3/44
• Note:
(1) A vector space consists of four entities:
a set of vectors, a set of scalars, and two operations
V：nonempty set
c：scalar
+ (u, v)  u  v vector addition
• . (c, u)  cu scalar multiplication

V , ,  is then called a vector space

(2) V  0 : zero vector space

4/44
• Examples of vector spaces:

(1) n-tuple space: Rn

(u1 , u 2 , , u n )  (v1 , v2 ,  , vn )  (u1  v1 , u 2  v2 ,  , un  vn ) vector addition
k (u1 , u2 , , un )  (ku1 , ku2 ,, kun ) scalar multiplication

(2) Matrix space: V  Mmn(the set of all m×n matrices with real values)
Exp: ：(m = n = 2)
u11 u12  v11 v12  u11  v11 u12  v12 
u u   v v   u  v u  v  vector addition
 21 22   21 22   21 21 22 22 
u11 u12  ku11 ku12 
k    scalar multiplication
u u ku
 21 22   21 ku22 
5/44
(3) n-th degree polynomial space: V  Pn (x)
(the set of all real polynomials of degree n or less)

p ( x)  q( x)  (a0  b0 )  (a1  b1 ) x    ( an  bn ) x n

kp( x)  ka0  ka1 x    kan x n

(4) Function space: V  c(, ) (the set of all real-

valued continuous functions defined on the entire real line.)

( f  g )( x)  f ( x)  g ( x)
(kf )( x)  kf ( x)

6/44
( f  g )( x)  f ( x)  g ( x) (kf )( x)  kf ( x)

7/44
 Theorem: (Properties of scalar multiplication)
Let v be any element of a vector space V, and let c be any
scalar. Then the following properties are true.
(1) 0v  0
(2) c0  0
(3) If cv  0, then c  0 or v  0
(4) (1) v   v

8/44
 Ex 6: The set of all integer is not a vector space.
Pf: 1  V , 12  R
( 12 )(1)  12  V (it is not closed under scalar multiplication)
  noninteger
scalar

integer

 Ex 7: The set of all second-degree polynomials is not a vector space.

Pf: Let p ( x)  x 2 and q ( x)   x 2  x  1

 p( x)  q( x)  x  1 V
(it is not closed under vector addition)
9/44
• Ex 8:
V=R2=the set of all ordered pairs of real numbers defined as:
- vector addition: (u1 , u2 )  (v1 , v2 )  (u1  v1 , u2  v2 )
- scalar multiplication: c(u1 , u2 )  (cu1 ,0)
---------Verify V is not a vector space.

Sol:
1(1, 1)  (1, 0)  (1, 1) Condition (10) is not satisfied
the set (together with the two given operations) is
not a vector space

10/44
11/44
 Vector Subspaces
• Subspace:

If (V ,,) : a vector space

W  
 : a nonempty subset
W V
(W ,,) ：a vector space (under the operations of addition and
scalar multiplication defined in V)
 Then W is a subspace of V
 Trivial subspace:
Every vector space V has at least two subspaces.
(1) Zero vector space {0} is a subspace of V.
(2) V is a subspace of V.
12/44
• Theorem: (Test for a subspace)

If W is a nonempty subset of a vector space V, then W is

a subspace of V if and only if the following conditions hold.

(1) If u and v are in W, then u+v is in W (closed under addition).

(2) If u is in W and c is any scalar, then cu is in W (closed under

multiplication by a scalar).

13/44
 Ex: Subspace of R2
(1) 0 0  0, 0 
(2) Lines through the origin
(3) R 2

• Ex: Subspace of R3

(1) 0 0  0, 0, 0 

(2) Lines through the origin
(3) Planes through the origin
(4) R 3

14/44
 Ex 2: (A subspace of M2×2)
Let W be the set of all 2×2 symmetric matrices. Show that
W is a subspace of the vector space M2×2, with the standard
operations of matrix addition and scalar multiplication.

Sol:
W  M 22 M 22 : vector sapces
Let A1, A2  W ( A1T  A1, A2T  A2 )
A1  W, A2  W  ( A1  A2 )T  A1T  A2T  A1  A2 ( A1  A2 W )
k  R , A  W  (kA)T  kAT  kA ( kA W )
W is a subspace of M 22
15/44
 Ex 3: (The set of singular matrices is not a subspace of M2×2)
Let W be the set of singular matrices of order 2. Show that
W is not a subspace of M2×2 with the standard operations.

Sol:
1 0 0 0 
A  W , B    W
0 0  0 1 

1 0
A B    W
0 1  singular matrices
The matrices are known to be singular if their determinant
equal to the zero
W2 is not a subspace of M 22

16/44
 2
 Ex 4: (The set of first-quadrant vectors is not a subspace of R )
Show that W  {( x1 , x2 ) : x1  0 and x2  0} , with the standard
operations, is not a subspace of R2.
(not closed under scalar
multiplication)
Sol:
Let u  (1, 1)  W
  1u   11, 1   1,  1  W

W is not a subspace of R 2

17/44
 Spanning Sets and Linear Independence
 Linear combination:
A vector v in a vector space V is called a linear combination of
the vectors u1,u 2 , ,u k in V if v can be written in the form

v  c1u1  c2u 2    ck u k c1,c2 , ,ck : scalars

18/44
 Ex 2-3: (To find linear combination)
v1  (1,2,3) v 2  (0,1,2) v 3  (1,0,1)
Prove (a) w  (1,1,1) is a linear combination of v1 , v 2 , v 3
(b) w  (1,2,2) is not a linear combination of v1 , v 2 , v 3
Sol:
(a) w  c1 v1  c2 v 2  c3 v 3
1,1,1  c1 1,2,3  c2 0,1,2  c3  1,0,1
 (c1  c3 , 2c1  c2 , 3c1  2c2  c3 )

c1  c3 1
 2c1  c2 1
3c1  2c2  c3 1
19/44
 1 0  1 1 1 0  1 1 
  Guass  Jordan Elimination
 2 1 0 1      
 0 1 2  1
 
3 2 1 1 0 0 0 0 

 c1  1  t , c2  1  2t , c3  t

(this system has infinitely many solutions)

t 1
 w  2 v1  3 v 2  v 3

20/44
(b)
w  c1 v1  c2 v 2  c3 v 3

1 0  1 1  1 0  1 1 
  GuassJordanElimination
 2 1 0  2 
 0 1 2  4 
 
3 2 1 2  0 0 0 7 

 this system has no solution ( 0  7)

 w  c1 v1  c2 v 2  c3 v 3

21/44
 the span of a set: span (S)
If S={v1, v2,…, vk} is a set of vectors in a vector space V,
then the span of S is the set of all linear combinations of
the vectors in S,

span(S ) c1 v1  c2 v 2    ck v k ci  R

(the set of all linear combinations of vectors in S )
 a spanning set of a vector space:

If every vector in a given vector space can be written as a

linear combination of vectors in a given set S, then S is
called a spanning set of the vector space.

22/44
23/44
 Notes:
span ( S )  V
 S spans (generates) V
V is spanned (generated) by S
S is a spanning set of V

 Notes:
(1) span( )  0
(2) S  span( S )
(3) S1 , S 2  V
S1  S 2  span( S1 )  span( S 2 )

24/44
 Linear Independence (L.I.) and Linear Dependence (L.D.):

S  v1 , v 2 , , v k  : a set of vectors in a vector space V

c1 v1  c2 v 2    ck v k  0

(1) If the equation has only the trivial solution (c1  c2    ck  0)

then S is called linearly independent.
(2) If the equation has a nontrivial solution (i.e., not all zeros),
then S is called linearly dependent.

25/44
 Notes:
(1)  is linearly independent
(2) 0  S  S is linearly dependent.
(3) v  0  v is linearly independent
(4) S1  S 2
S1 is linearly dependent  S 2 is linearly dependent

S 2 is linearly independent  S1 is linearly independent

26/67
 Ex 8: (Testing for linearly independent)
Determine whether the following set of vectors in R3 is L.I. or L.D.
S  1, 2, 3, 0, 1, 2 ,  2, 0, 1
v1 v2 v3
Sol: c1  2c3  0
c1 v1  c2 v 2  c3 v 3  0  2c1  c2  0
3c1  2c2  c3  0
1 0  2 0  1 0 0 0
 2 1 0 0 Gauss
   0 1 0 0
- Jordan Elimination

3 2 1 0 0 0 1 0

 c1  c2  c3  0 only the trivial solution 
 S is linearly independent
27/44
• Ex 10: (Testing for linearly independent)
Determine whether the following set of vectors in 2×2
matrix space is L.I. or L.D.

2 1 3 0 1 0 
S    ,  ,  
0 1 2 1 2 0 
v1 v2 v3
Sol:
c1v1+c2v2+c3v3 = 0

2 1 3 0 1 0   0 0 
c1    c2    c3    
 0 1  2 1   2 0   0 0 

28/44
 2c1+3c2+ c3 = 0
c1 =0
2c2+2c3 = 0
c1+ c2 =0
  2 3 1 0 1 0 0 0
1 0 0 0  0
  1 0 0
Gauss
   
- Jordan Elimination

0 2 2 0 0 0 1 0
   
1 1 0 0  0 0 0 0

 c1 = c2 = c3= 0 (This system has only the trivial solution.)

 S is linearly independent.
29/44
 Basis and Dimension
• Basis:
Linearly
V：a vector space Generating
Bases Independent
Sets(span V)
Sets
S ={v1, v2, …, vn}V

(a ) S spans V (i.e., span(S) = V )

If  Intersection-‫التقاطع‬
(b) S is linearly independent
 Then S is called a basis for V

 Notes:
(1) Ø is a basis for {0}
(2) the standard basis for R3:
{i, j, k} i = (1, 0, 0), j = (0, 1, 0), k = (0, 0, 1)
30/44
 n
(3) the standard basis for R :
{e1, e2, …, en} e1=(1,0,…,0), e2=(0,1,…,0), en=(0,0,…,1)
Ex: R4 {(1,0,0,0), (0,1,0,0), (0,0,1,0), (0,0,0,1)}

(4) the standard basis for mn matrix space:

{ Eij | 1im , 1jn }
Ex: for M22 2 2 matrix space:
1 0 0 1 0 0 0 0 
  ,  ,  ,  
0 0 0 0 1 0 0 1 
(5) the standard basis for Pn(x):
{1, x, x2, …, xn}
Ex: P3(x) {1, x, x2, x3}
31/44
• Thm 4.9: (Uniqueness of basis representation)
If S v1,v2,,vn is a basis for a vector space V, then every
vector in V can be written in one and only one way as a linear
combination of vectors in S.

Proof:
 1. span(S) = V
 S is a basis  
 2. S is linearly independent
 span(S) = V Let v = c1v1+c2v2+…+cnvn
v = b1v1+b2v2+…+bnvn
 0 = (c1–b1)v1+(c2 – b2)v2+…+(cn – bn)vn

 S is linearly independent
 c1= b1 , c2= b2 ,…, cn= bn (i.e., uniqueness)
32/ 44
• Thm 4.10: (Basis and linear dependence)
 
If S  v1, v2,, vn is a basis for a vector space V, then every
set containing more than n vectors in V is linearly dependent.

Proof:
Let S1 = {u1, u2, …, um} , m > n
 span ( S )  V
u1  c11 v1  c21 v 2    cn1 v n
uiV  u 2  c12 v1  c22 v 2    cn 2 v n

u m  c1m v1  c2 m v 2    cnm v n

33/44
Let k1u1+k2u2+…+kmum= 0
 d1v1+d2v2+…+dnvn= 0 (where di = ci1k1+ci2k2+…+cimkm)
 S is L.I.
 di=0 i i.e. c11k1  c12 k 2    c1m k m  0
c21k1  c22 k 2    c2 m k m  0

cn1k1  cn 2 k 2    cnm k m  0

 According to Thm 1.1: If the homogeneous system has fewer

equations than variables, then it must have infinitely many solution.

m > n  k1u1+k2u2+…+kmum = 0 has nontrivial solution

 S1 is linearly dependent
34/ 44
• Theorem: (Number of vectors in a basis)
If a vector space V has one basis with n vectors, then every
basis for V has n vectors. (All bases for a finite-dimensional
vector space has the same number of vectors.)

Proof:
S ={v1, v2, …, vn}
two bases for a vector space
S'={u1, u2, …, um}

S is a basis  Thm.4.10 
  n  m
S ' is L.I.  
n  m
S is L.I.  Thm . 4 .10
  n  m
S ' is a basis  
35/44
• Finite dimensional:
A vector space V is called finite dimensional,
if it has a basis consisting of a finite number of elements.

 Infinite dimensional:
If a vector space V is not finite dimensional,
then it is called infinite dimensional.
 Dimension:
The dimension of a finite dimensional vector space V is
defined to be the number of vectors in a basis for V.
V: a vector space S: a basis for V

symbol: dim(V) = #(S) (the number of vectors in S)

36/44
 Notes: dim(V) = n
Linearly
(1) dim({0}) = 0 = #(Ø) Generating
Bases Independent
Sets Sets

(2) dim(V) = n , SV #(S) > n #(S) = n #(S) < n

S：a generating set  #(S)  n

S：a L.I. set  #(S)  n
S：a basis  #(S) = n

(3) dim(V) = n , W is a subspace of V  dim(W)  n

37/44
• Exp:
(1) Vector space Rn  basis {e1 , e2 ,  , en}
 dim(Rn) = n
(2) Vector space Mmn  basis {Eij | 1im , 1jn}
 dim(Mmn)=mn
(3) Vector space Pn(x)  basis {1, x, x2,  , xn}
 dim(Pn(x)) = n+1
(4) Vector space P(x)  basis {1, x, x2, }
 dim(P(x)) = 

38/ 44
• Ex 9: (Finding the dimension of a subspace)
(a) W={(d, c–d, c): c and d are real numbers}
(b) W={(2b, b, 0): b is a real number}

Sol: (Note: Find a set of L.I. vectors that spans the subspace)
(a) (d, c– d, c) = c(0, 1, 1) + d(1, – 1, 0)
 S = {(0, 1, 1) , (1, – 1, 0)} (S is L.I. and S spans W)
 S is a basis for W
 dim(W) = #(S) = 2
(b)  2b, b,0   b2,1,0 
 S = {(2, 1, 0)} spans W and S is L.I.
 S is a basis for W
 dim(W) = #(S) = 1
39/ 44
 Ex 10: (Finding the dimension of a subspace)
Let W be the subspace of all symmetric matrices in M22.
What is the dimension of W?

Sol:
a b  
W    a , b, c  R 
b c  
a b  1 0   0 1   0 0 
   a   b   c 
 b c   0 0   1 0   0 1 
1 0 0 1 0 0 
 S    ,  ,   spans W and S is L.I.
  0 0  1 0   0 1  
 S is a basis for W  dim(W) = #(S) = 3
40/ 44
 • Theorem: (Basis tests in an n-dimensional space)
Let V be a vector space of dimension n.
(1) If S  v1, v2 ,, vn  is a linearly independent set of
vectors in V, then S is a basis for V.
(2) If S  v1, v2 ,, vn  spans V, then S is a basis for V.

dim(V) = n
Imp: If we have a space V of dimension
n, and a set of vectors S of number equal
n, then for the set of vectors S to be a Generating Linearly
Basis of V, it is sufficient to show that S is Sets Bases Independent
L.I. or that it spans V. Sets

#(S) > n #(S) < n

#(S) = n 41/44
42/44
43/43
44/44