Setup and goals

MATH214
• Discover common methods to prove results
Linear Algebra
• Understand the structure of a proof

• Participate to in-class discussions

Discussion 4
• Challenge yourself with a guided problem
Manuel — JI (Spring 2025)

Matrix of a linear map

Let f : V → W be a linear map, and BV = {v1 , · · · , vn } and BW = {w1 , · · · , wp } be two bases for
V and W , respectively. Determine the matrix A of the linear map f .

Method
For any j ∈ J1, nK, the jth column is composed of the coordinates f (vj ) in the basis Bw .

Example
Determine the matrix of f : R2 [X ] → R3 [X ], such that f (P) = XP. Use the canonical bases of
R2 [X ] and R3 [X ].
It is clear that f is a linear map, and using the canonical( basis)of R2 [X ] we have f (1) = X ,
000
f (X ) = X 2 , and f (X 2 ) = X 3 . Hence we obtain the matrix 10 01 00 .
001

Ex. 1 — Matrix of an endomorphism

In the canonical basis of R2 [X ], determine the matrix of the endomorphism f of R2 [X ] defined such that
f (P) = P ′ .

Ex. 2 — Endomorphism of M2 (R)

{ ( ) ( ) ( ) ( )}
We define the canonical basis B of M2 (K) to be E1,1 = 10 00 , E1,2 = 00 10 , E2,1 = 01 00 , E2,2 = 00 01 .
( )
Let f be the endomorphism of M2 (K) such that f (M) = AM, where A = 01 10 . Determine the matrix
of f in B.

Ex. 3 — Change of basis matrix

Let B1 = {(3, 1, 4), (5, 3, 2), (1, −1, 7)} and B2 = {(1, 2, 1), (2, 3, 3), (3, 7, 1)} be two subsets of R3 .

1. Show that B1 and B2 are two bases of R3 .

2. Determine the change of matrix from B1 to B2 .

 Inverse of a matrix
Prove that a square matrix A ∈ Mn (K) has an inverse and calculate it.

Method
Linear map. Consider the linear map f associated to A, prove that it is bijectif, determine f −1 ,
and deduce A−1 from it.
Create a system of equations. Let {E1 , · · · En } be the canonical basis of Mn,1 and C1 , · · · Cn be
the columns of A. Express the Ci , 1 ≤ i ≤ n in terms of the Ei , 1 ≤ i ≤ n, then solve the system
of equations in the unknowns Ei , 1 ≤ i ≤ n, and finally deduct A−1 .
Use a matrix equation. Form an equation involving In and A. Isolate In on one side and on the
other factorize by A.
Examples
( 121
)
Linear map. Prove that A = 131 is invertible and calculate its inverse.
−1 4 0
We can read A as the matrix of an endomorphism f in the canonical basis B = {e1 , e2 , e3 } of R3 .
If we note f (ei ) = ui , 1 ≤ i ≤ 3 then we have u1 = e1 + e2 − e3 , u2 = 2e1 + 3e2 + 4e3 , and
u3 = e1 + e2 . By linear combinations
( ) get e1 = −4u1 − u2 + 7u3 , e2 = 4u1 + u2 − 6u3 , and
we
−4 4 −1
e3 = u3 − u1 . Hence A−1 = −1 1 0 .
7 −6 1
( 121
)
Create a system of equations. Show that 110 is invertible and calculate its inverse.
−2 0 1
Let B = {E1 , E2 , E3 } be the canonical basis of M3,1 (R) and Ci , 1 ≤ i ≤ 3 be the columns of A.
Then we have
  

 C = E1 + E2 − 2E3 
 E = C2 − 2E1 
 E = C1 − C2 + 2C3

 1 
 2 
 1
C2 = 2E1 + E2 ⇔ E3 = C3 − E1 ⇔ E2 = −2C1 + 3C2 − 4C3 .

 
 


C3 = E1 + E2 
C1 = E1 + (C2 − 2E1 ) − 2(C3 − E1 ) 
E3 = −C1 + C2 − C3

( )
1 −2 −1
This proves that A is invertible and has inverse A−1 = −1 3 1 .
2 −4 −1

Create a matrix equation. Let A ∈ M3 (K) such that 4A2 −3A−I3 = 0. Show that A is invertible
and determine its inverse.
Based on the given equation we observe that

A(4A − 3I3 ) = I3
.
(4A − 3I3 )A = I3

It is then clear that A is invertible and A−1 = 4A − 3I3 .

Ex. 4 — Inverse of a square matrix

( )
Let n ∈ N∗ and A = min(i, j)1≤i,j≤n ∈ Mn (K).

1. Express A in a non-compact form.

2. Show that A is invertible and determine its inverse.

Ex. 5 — Inverse of simple matrices
(1 1 1) (1 1 0)
Show that 0 1 1 and 1 1 1 are invertible and determine their inverse.
001 011

Ex. 6 — Inverse of a polynomial of square matrices

Let n ∈ N∗ and A ∈ Mn (K) such that A5 + A = In . Determine the inverse of A2 + A + In .
Hint: perform the Euclidean division of X 5 + X − 1 by X 2 + X + 1.

Guided exercise
Let G = ⟨V , E ⟩ be a graph, where V = {v1 , · · · , vn } and E are the sets of the vertices and edges,
respectively. We define ai,j to be 1 if (vi , vj ) ∈ E and 0 otherwise. The matrix A = (ai,j )1≤i,j≤n is
called the adjacency matrix of G .

Questions
1. Prove that the number of paths of length k connecting vi and vj is the element at position
(i, j) in Ak .
2. Let Mn be an adjacency matrix for a complete graph Kn . Calculate Mn2 .
3. Let Jn ∈ Mn (R) be the matrix only composed of ones. Show that for any k ∈ N∗ ,

(n − 1)k − (−1)k
Mnk = Jn + (−1)k In .
n

Getting started
Before starting, ensure you are fully familiar with the notions of simple and complete graphs.
1. Reason by induction on the length of the paths.
2. Noticing that Kn is complete, decompose Mn using Jn and In .
3. Reason by induction on k.