Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd
Upload
43 views

4 Pictures of The Same Thing: Picture (I) : Systems of Linear Equations

The document discusses four different ways of representing the same system of linear equations: (i) As a system of equations, (ii) as linear combinations of vectors, (iii) as a matrix equation, and (iv) as a linear transformation. Each representation provides a different perspective on questions of existence and uniqueness of solutions to the system.

Uploaded by

thezackattack
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PDF, TXT or read online on Scribd
43 views

4 Pictures of The Same Thing: Picture (I) : Systems of Linear Equations

The document discusses four different ways of representing the same system of linear equations: (i) As a system of equations, (ii) as linear combinations of vectors, (iii) as a matrix equation, and (iv) as a linear transformation. Each representation provides a different perspective on questions of existence and uniqueness of solutions to the system.

Uploaded by

thezackattack
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

4 Pictures of the same thing

Most of Chapter 1 has been concerned with looking at the same object (a system
of linear equations) in several dierent ways. Heres a summary of whats going
on!
Picture (i): Systems of linear equations
A general system of linear equations is written as
_

_
a
11
+ + a
1n
= b
1
.
.
.
a
m1
+ + a
mn
= b
m
We often eliminate the + and = signs by writing it as an augmented matrix
(i)
_

_
a
11
. . . a
1n
b
1
.
.
.
.
.
.
.
.
.
.
.
.
a
m1
. . . a
mn
b
m
_

_
Picture (ii): Linear combinations of vectors
Vectors in R
d
A d-dimensional vector v is an object with a length and a direction. By taking
v
1
, . . . , v
d
to be its components in the x
1
, . . . , x
d
directions we can write any v
as
v =
_

_
v
1
.
.
.
v
d
_

_
We can identify any d-dimensional vector v with a unique point
(v
1
, . . . , v
d
) R
d
and hence we often talk about d-dimensional vectors as elements of R
d
.
When we add two vectors we add each component seperately for example
_
_
1
2
3
_
_
+
_
_
1
0
1
_
_
=
_
_
2
2
2
_
_
1
When we multiply a vector by a scalar we multiply each component by the
scalar. For example
3
_
_
1
2
3
_
_
=
_
_
3
6
9
_
_
Spans and linear independence
We dene the span of a set of vectors a
1
, . . . , a
n
to be the set of all vectors v
that can be written as a linear combination of these vectors, that is to say
v =
n

j=1
a
j
x
j
for real numbers x
1
, . . . , x
n
We denote the span by
sp{a
1
, . . . , a
n
}
A collection of vectors a
1
, . . . , a
n
is said to be linearly independent if the
only solution to
n

j=1
a
j
x
j
= 0
is given by x
1
= = x
n
= 0.
Writing the system (i) with vectors
Returning to our system (i), dene m-dimensional vectors
a
j
=
_

_
a
1j
.
.
.
a
mj
_

_ for j = 1, . . . , n
and the vector
b =
_

_
b
1
.
.
.
b
m
_

_
Then our system can be written as
(ii) a
1
x
1
+ +a
n
x
n
= b
2
Picture (iii): Matrix equations
Matrices and applying matrices to vectors
We dene an m n matrix A to be a collection of m rows and n columns of
numbers. For example
A =
_

_
a
11
. . . a
1n
.
.
.
.
.
.
.
.
.
a
m1
. . . a
mn
_

_
Given an n-dimensional vector
x =
_

_
x
1
.
.
.
x
n
_

_
and taking the columns of the matrix to be the m-dimensional vectors a
1
, . . . , a
n
A =
_
a
1
. . . a
n

we can apply the matrix to the vector by


Ax =
n

j=1
a
j
x
j
A way to think about matrices applied to vectors is to set
y = Ax
then each coecient a
ij
is the amount of the component of the input x
j
in
the component of the output y
i
. Each column of the matrix corresponds to an
input variable x
j
and each row to an output variable y
i
.
Writing the system (i) with matrices
By taking A as above we note that the left hand side of (ii) is simply Ax and
hence we can re-write the system (i) as
(iii) Ax = b
where b is as above.
Picture (iv): Linear transformations
We say a map T : R
n
R
m
is linear if for any two vectors u, v R
n
and any
scalar R
T(u +v) = T(u) + T(v)
3
A fact that we will use a lot is that for any such linear map T we can nd a
matrix A such that
T(x) = Ax
We call the set of all possible outputs of the map T the range of T. We say
T is onto if the range of T is the whole of R
m
.
We say the map T is one-to-one if there are no non-trivial (non-zero) solu-
tions x to T(x) = 0.
Writing system (i) with linear maps
Given our linear system above and the coecient matrix A we can dene a
linear map by
T(x) = Ax
The system then becomes the equation
(iv) T(x) = b
Existence and uniqueness questions
For each way of writing the system (i) we have a corresponding question about
existence and uniqueness.
The following questions are all equivalent using the 4 dierent pictures of
system (i).
1. Is the system (i) consistent of all choices of b
1
, . . . , b
m
?
2. In (ii), does sp{a
1
, . . . , a
n
} = R
m
?
3. In (iii), does the inhomogeneous equation Ax = b have a solution for all
vectors b R
m
?
4. In (iv), is the linear map T onto?
Similarly, the following questions are all equivalent using the 4 dierent
pictures of system (i).
1. If the system (i) is consistent, is the solution unique?
2. In (ii), are the vectors a
1
, . . . , a
n
linearly independent?
3. In (iii), is the only solution to the homogeneous equation Ax = 0 the
trivial solution?
4. In (iv), is the linear map T one-to-one?
4

You might also like