Lecture 07 recap

1) How elementary row operations changes the

determinant of a matrix.
2) Using elementary row operations to find the
determinant of a matrix.
3) A is invertible if and only if det( A)  0.
4) det(cA), det( AB) and det( A1).
5) Definition of adjoint of a matrix. Inverse in terms
of adjoint.
6) Cramer's rule.
Lecture 08
Euclidean n-spaces
Linear combinations and
linear spans
How do we represent vectors
Geometrically, a vector is represented by a directed
line segment (or arrow).

direction of the arrow

length of the arrow (magnitude)

Two vectors u and v are said to be equal if they

have the same length and direction.

u v
 Vector operations
1) Addition: u  v

v vu
u
u
uv v
u+ v  v + u

2) Negative of u:  u

u u change in direction,
no change in magnitude
 Vector operations
3) Difference: u  v (same as u  (v))

v
uv v
u

4) Scalar multiple: ku (k  )

u 1
u 2u (1.2)u
2
 Coordinate systems (2-space)
Position u with its initial point at (0, 0)
y

(u1 , u2 )
u  (u1 , u2 ) u2
u
components of u
(0, 0) u1 x

xy  plane
 Coordinate systems (3-space)
Position u with its initial point at (0, 0, 0)

u3 (u1 , u2 , u3 )
u  (u1 , u2 , u3 )
u
components of u u2
(0, 0, 0) y
u1
x xyz  space
 Addition and scalar multiplication
1) Addition: add component-wise.

(u1 , u2 )  (v1 , v2 )  (u1  v1 , u2  v2 )

(u1 , u2 , u3 )  (v1 , v2 , v3 )  (u1  v1 , u2  v2 , u3  v3 )

2) Scalar multiplication: multiply to each component.

u  (u1 , u2 ), k  ku  (ku1 , ku2 )

u  (u1 , u2 , u3 ), k  ku  (ku1 , ku2 , ku3 )

Definitions (n-vectors and operations)
n-vector: (u1 , u2 ,..., ui ,..., un )

u1 , u2 ,..., un are real numbers.

u1 : 1st component (or 1st coordinate) of the vector

ui : ith component (or ith coordinate) of the vector

u  (u1 , u2 ,..., un ) v  (v1 , v2 ,..., vn )

Equality: u  v if and only if

ui  vi for all i  1, 2,..., n.

Definitions (n-vectors and operations)
u  (u1 , u2 ,..., un ) v  (v1 , v2 ,..., vn )

Addition: u  v  (u1  v1 , u2  v2 ,..., un  vn )

Scalar multiplication: c  , cu  (cu1 , cu2 ,..., cun )

Negative:  u  (u1 , u2 ,..., un )

Subtraction: u  v  (u1  v1 , u2  v2 ,..., un  vn )

Zero vector: 0  (0, 0,..., 0)

 Example (vector operations)
u  (3, 4, 5,1) v  (1, 0,1, 2)

u  2v 

2u  3v 

If w  (0, 0, 0, 0, 0), what is u  2 w?

 Vectors and matrices
Identifying an n-vector (u1 ,u2 ,...,un ) with:

1 n matrix  u1 u2 ... un  (row vector)

or

 u1 
u 
n 1 matrix  2  (column vector)
 : 
u 
 n
Properties of vector operations
Let u, v , w be n-vectors and a , b be real numbers.

1) u  v  v  u 5) a(bu)  (ab)u

2) u  (v  w)  (u  v)  w 6) a(u  v)  au  av

3) u  0  u  0 + u 7) (a  b)u  au  bu

4) u  (u)  0 8) 1u  u

Proof: Not discussed here.

 Definition (Euclidean n-space)
Euclidean n-space, denoted by n , is the set of all
n-vectors (u1 , u2 ,..., un ) where ui , i  1,..., n, is a real number.

u  (u1 ,u2 ,...,un )  n

if and only if u1 ,..., un  .

Note:
n
1) For any positive integer n, is a set.
2) How many vectors does n
contain? Infinitely many!
2 3
3) Do and have any vector in common? No!
(2, 3) (1.5, 3 , 0)
 Example (Subsets of Rn)
S  {(u1 , u2 , u3 )|u1  0 and u2  u3 }

3
S contains vectors (u1 ,u2 ,u3 ) from such that

u1  0 and u2  u3

S is a subset of 3
(0,1, 1)  S (0, 0, 0)  S (1, 0, 0)  S

We can also write S as

S  {(0, a , a)| a  }
 Example (Subsets of Rn)
S  {(w, x , y , z)| w  x  y  z  0, 2w  x  y  2 z  1}

S contains vectors (w, x , y , z) from 4 such that w, x , y , z

satisfies
w  x  y  z  0

2w  x  y  2 z  1

S is the solution set of the above linear system.

4
S is a subset of .
Note that the solution set of a linear system involving
n
n variables will be a subset of .
 Example (Subsets of Rn)
S  {(w, x , y , z)| w  x  y  z  0, 2w  x  y  2 z  1}
(Implicit)
Solving the linear system, we have a general solution

w  13  t
x We are now able to rewrite
  13  s
 the set S in another way:
y  s
 z  t s ,t 

S  {(13  t , 13  s , s , t )| s , t  } (Explicit)
 Example (Subsets of Rn)
2
Lines in
Recall that the equation ax  by  c in two variables x , y
2
represents a line in .
Solving
(Implicit) representation:

{( x , y)| ax  by  c}, a , b  not both zero.

(Explicit) representation:

c  bt c  at
{( , t )|t  }, if a  0; {(t , )|t  }, if b  0.
a b
 Example (Subsets of Rn)
3
Planes in
Recall that the equation ax  by  cz  d in three variables
3
x , y , z represents a plane in .
(Implicit) representation:

{( x , y , z)| ax  by  cz  d }, a , b , c  not all zero.

What about explicit representation?

Solving ax  by  cz  d . One equation, 3 unknowns.

 Example (Subsets of Rn)
3
Planes in
Recall that the equation ax  by  cz  d in three variables
3
x , y , z represents a plane in .

Explicit representations: d  bs  ct
{( , s , t )| s , t  }, if a  0;
a
d  as  ct
{(s , , t )| s , t  }, if b  0;
b
d  as  bt
{(s , t , )| s , t  }, if c  0;
c
 Example (Subsets of Rn)
3
How do we represent lines in ?

3
A line in is usually represented explicitly.
A point and a
So what do
direction...or
we need?
two points.
 Example (Subsets of Rn)
3
A line in is represented by the set

{ (a0 , b0 , c0 ) t (a , b , c) |t  }

(point on the line) (a0 , b0 , c0 )

(a0 , b0 , c0 )  2(a , b, c)
(a0 , b0 , c0 )  (a , b , c)

(a0 , b0 , c0 )  (a , b , c)
(a , b, c) (direction of the line)
origin
 Example (Subsets of Rn)
3
A line in is represented by the set

{(a0 , b0 , c0 )  t (a , b , c)|t  }  {(a0  ta , b0  tb , c0  tc)|t  }

Remember:
3
A line in cannot be represented by a
2
single linear equation like in .
 Notation
S is a finite set. We use S to denoted the number
of elements in S .

S1  {1, 2, 3}, S2  {(1, 2, 3)}, S3  {(1, 2, 3),(2, 3, 4)}

S1  3, S2  1, S3  2
Definition (Linear Combination)
Consider u  (1, 2, 1), v  (0, 2, 5).
2u  3v  (2,10,13) u  2v  (1, 2, 11)
(2,10,13) and (1, 2, 11) are both linear combinations
of u and v.
n
Let u1 , u2 ,..., uk be vectors in .
For any real numbers c1 , c 2 ,..., ck , the vector
c1u1  c 2 u2  ...  ck uk

is a linear combination of u1 , u2 ,..., uk .

Example (Linear Combination)
Consider u  (1, 2, 1), v  (0, 2, 5), w  (1, 0, 2).

Question: Compute the linear combination 2u  3v  w

Answer: This is simple.

2u  3v  w 
Example (Linear Combination)
Consider u  (1, 2, 1), v  (0, 2, 5), w  (1, 0, 2).

Question: Is (0, 4,8) a linear combination of u, v , w ?

Answer: We need to check whether there are real numbers

a , b, c such that
au  bv  cw  (0, 4,8)
a(1, 2, 1)  b(0, 2, 5)  c(1, 0, 2)  (0, 4,8)

How to check?
Example (Linear Combination)
Consider u  (2,1, 3), v  (1, 1, 2), w  (3, 0, 5).
Question: Is (3, 3, 4) a linear combination of u, v , w ?

au  bv  cw  (3, 3, 4)

a(2,1, 3)  b(1, 1, 2)  c(3, 0, 5)  (3, 3, 4)

2a  b  3c  3  2 1 3 3
  1 1 0 3 
a  b  3
 
3a  2b  5c  4  3 2 5 4
  
Example (Linear Combination)
Consider u  (2,1, 3), v  (1, 1, 2), w  (3, 0, 5).
Question: Is (3, 3, 4) a linear combination of u, v , w ?

au  bv  cw  (3, 3, 4) a(2,1, 3)  b(1, 1, 2)  c(3, 0, 5)  (3, 3, 4)

 2 1 3 3  2 1 3 3
 1 1 0 3  Gaussian 0 3 3 3 
  Elimination  2 2 2

 3 2 5 4 0 0 0 0
   

Is the linear system consistent? Are the solutions unique?

Example (Linear Combination)
Consider u  (2,1, 3), v  (1, 1, 2), w  (3, 0, 5).
Question: Is (3, 3, 4) a linear combination of u, v , w ?

au  bv  cw  (3, 3, 4) a(2,1, 3)  b(1, 1, 2)  c(3, 0, 5)  (3, 3, 4)

a  2  t (a , b, c)  (2, 1, 0)

b  1  t 2(2,1, 3)  (1, 1, 2)  0(3, 0, 5)  (3, 3, 4)
c  t t

(a , b , c)  (1, 2,1)
(2,1, 3)  2(1, 1, 2)  (3, 0, 5)  (3, 3, 4)
Example (Linear Combination)
Consider u  (2,1, 3), v  (1, 1, 2), w  (3, 0, 5).
Question: Is (1, 2, 4) a linear combination of u, v , w ?

au  bv  cw  (1, 2, 4)

a(2,1, 3)  b(1, 1, 2)  c(3, 0, 5)  (1, 2, 4)

2a  b  3c  1 2 1 3 1
  1 1 0 2 
a  b  2
 
3a  2b  5c  4  3 2 5 4
  
Example (Linear Combination)
Consider u  (2,1, 3), v  (1, 1, 2), w  (3, 0, 5).
Question: Is (1, 2, 4) a linear combination of u, v , w ? No!

au  bv  cw  (1, 2, 4) a(2,1, 3)  b(1, 1, 2)  c(3, 0, 5)  (1, 2, 4)

2 1 3 1  2 1 3 1
 1 1 0 2  Gaussian 0 3 3 3 
  Elimination  2 2 2

 3 2 5 4  0 0 0 3
   

Is the linear system consistent?

 Example (Linear Combination)
Consider e1  (1, 0, 0, 0), e2  (0,1, 0, 0), e3  (0, 0,1, 0), e4  (0, 0, 0,1)
(1, 2, 3, 4) 

(3, 13 , 0, 2) 

4
Any (w, x , y , z) in :

Every vector u  (w, x , y , z) in 4

is a linear combination
of e1 , e2 , e3 , e4 .
 Example (Linear Combination)
Consider u  (1, 2, 1), v  (0, 2, 5), w  (1, 0, 2).
3
Question: Is every vector in a linear combination of u, v , w ?
 Discussion (Linear Combination)
Consider u  (1, 2, 1), v  (0, 2, 5), w  (1, 0, 2).
How many different linear
combinations of u, v and w
Quite a bit...
are there?

What if I put ALL different OMG...

linear combinations of
u, v and w into a set?
 Definition (Linear Span)
Let S  { u1 , u2 ,..., uk } be a set of vectors in n
.

The set of all linear combinations of u1 , u2 ..., uk ,

{c1u1  c 2 u2  ...  ck uk |c1 , c2 ,..., ck  }

is called the linear span of S (or linear span of u1 , u2 ,..., uk ).

This set is denoted by span(S ) or span{ u1 , u2 ,..., uk }.

 Example
Consider u  (2,1, 3), v  (1, 1, 2), w  (3, 0, 5).
Question: Is (3, 3, 4) a linear combination of u, v , w ? Yes!
So (3, 3, 4)  span{ u, v , w}

Question: Is (1, 2, 4) a linear combination of u, v , w ? No!

So (1, 2, 4)  span{ u, v , w}
 Example
S  {(1,1, 0),(2, 1,1)}.

span(S )  set of all linear combinations of (1,1, 0) and (2, 1,1)

Every vector in span(S ) is of the form

a(1,1, 0)  b(2, 1,1) where a , b are any real numbers.

So span(S )  { a(1,1, 0)  b(2, 1,1)| a , b  }

 Example
V  {(2a  b , a , 3b  a)| a , b  }
3
V is a subset of . Can V be written as a linear span?

(2a  b , a , 3b  a)
 a(2,1, 1)  b(1, 0, 3)

So V  { a(2,1, 1)  b(1, 0, 3)| a , b  }

 span{(2,1, 1),(1, 0, 3)}
 Example
Show that span{(1, 0,1),(1,1, 0),(0,1,1)}  3
.
We need to show that every vector in 3 can be
written as a linear combination of (1, 0,1),(1,1, 0),(0,1,1).

Show that span{(1,1,1),(1, 2, 0),(2,1, 3),(2, 3,1)}  3

.
We need to show that there is some vector in 3
that cannot be written as a linear combination of
(1,1,1),(1, 2, 0),(2,1, 3),(2, 3,1).
 End of Lecture 08
Lecture 09:
Linear combinations and linear spans (cont’d)
Subspaces (till end of Section 3.3)