Lecture 23 - The Inverse of a Matrix

Learning Objectives

Definition of Inverse

Definition. Given an \( n\times n \) matrix \( A \), the multiplicative inverse of \( A \) is a matrix \( A^{-1} \) such that \( A\cdot A^{-1} = A^{-1}\cdot A = I_n \).

Unlike the one-sided "inverses" we learned about in the previous lecture, \( A^{-1} \) must be an inverse that works when multiplying on the left and on the right. We also learned that a matrix must be square for such a "two-sided" inverse to exist. Even when a matrix is square, it is not guaranteed that an inverse exists.

Definition. If a square matrix \( A \) does not have an inverse, it is singular. A matrix that does have an inverse is called invertible or nonsingular.

Example 1. Let \( A = \begin{bmatrix} -1 & 1 & 0 \\ 2 & 0 & 1 \\ 0 & 1 & 1 \end{bmatrix} \) and \( B = \begin{bmatrix} 1 & 1 & -1 \\ 2 & 1 & -1 \\ -2 & -1 & 2 \end{bmatrix} \). Verify that \( A \) and \( B \) are inverses of one another.

We compute: \[ AB = \begin{bmatrix} -1 & 1 & 0 \\ 2 & 0 & 1 \\ 0 & 1 & 1 \end{bmatrix} \begin{bmatrix} 1 & 1 & -1 \\ 2 & 1 & -1 \\ -2 & -1 & 2 \end{bmatrix} = \begin{bmatrix} 1 & 0 & 0 \\ 0 & 1 & 0 \\ 0 & 0 & 1 \end{bmatrix} \] \[ BA = \begin{bmatrix} 1 & 1 & -1 \\ 2 & 1 & -1 \\ -2 & -1 & 2 \end{bmatrix} \begin{bmatrix} -1 & 1 & 0 \\ 2 & 0 & 1 \\ 0 & 1 & 1 \end{bmatrix} = \begin{bmatrix} 1 & 0 & 0 \\ 0 & 1 & 0 \\ 0 & 0 & 1 \end{bmatrix} \] Since both \( AB = I_3 \) and \( BA = I_3 \), we see that \( A \) and \( B \) are inverses of one another. \( \Box \)

Inverse of a \( 2\times 2 \) Matrix

In the case of a \( 2 \times 2\) matrix \( A \), we have a convenient way of determining whether \( A \) is invertible:

Definition. Let \( A = \begin{bmatrix} a & b \\ c & d \end{bmatrix} \), where \( a,b,c,d \in \mathbb R \). The determinant of \( A \) is \( \det A = ad-bc \).

Theorem. If \( A = \begin{bmatrix} a & b \\ c & d \end{bmatrix} \) and \( \det A \ne 0 \), then \( A \) is invertible and \( A^{-1} = \frac 1 {\det A} \begin{bmatrix} d & -b \\ -c & a \end{bmatrix} \).

Proof. We multiply to verify that \( A\cdot A^{-1} = I_2 \) and \( A^{-1}\cdot A = I_2 \): \[ A\cdot A^{-1} = \begin{bmatrix} a & b \\ c & d \end{bmatrix} \cdot \frac 1 {ad-bc} \begin{bmatrix} d & -b \\ -c & a \end{bmatrix} = \frac 1 {ad-bc} \begin{bmatrix} a & b \\ c & d \end{bmatrix} \begin{bmatrix} d & -b \\ -c & a \end{bmatrix} = \frac 1 {ad-bc} \begin{bmatrix} ad-bc & 0 \\ 0 & ad-bc \end{bmatrix} = \begin{bmatrix} 1 & 0 \\ 0 & 1 \end{bmatrix} \] \[ A^{-1}\cdot A = \frac 1 {ad-bc} \begin{bmatrix} d & -b \\ -c & a \end{bmatrix} \cdot \begin{bmatrix} a & b \\ c & d \end{bmatrix} = \frac 1 {ad-bc} \begin{bmatrix} ad-bc & 0 \\ 0 & ad-bc \end{bmatrix} = \begin{bmatrix} 1 & 0 \\ 0 & 1 \end{bmatrix}.\ \Box \]

Example 2. Let \( A = \begin{bmatrix} 7 & -3 \\ 4 & -2 \end{bmatrix} \). Find \( A^{-1} \).

Using the formula, we have \( \det A = (7)(-2)-(-3)(4) = -2 \). So, \( A^{-1} = \frac 1{-2} \begin{bmatrix} -2 & 3 \\ -4 & 7 \end{bmatrix} = \begin{bmatrix} 1 & -3/2 \\ 2 & -7/2 \end{bmatrix} \). \( \Box \)

Inverses and Matrix Equations

Invertible matrices are convenient because they allow us to easily solve matrix equations if we know the inverse of the matrix.

Theorem (Invertible Matrix Equations). If \( A \) is an invertible \( n\times n \) matrix, then for each \( \bbm b\in \mathbb R^n \) the equation \( A \bbm x = \bbm b \) has a unique solution \( \bbm x = A^{-1} \bbm b \).

Proof. We first show that \( \bbm x = A^{-1} \bbm b \) is a solution for \( A \bbm x = \bbm b \). Since \( A(A^{-1}\bbm b) = (AA^{-1})\bbm b = I_n \bbm b = \bbm b \), the vector \( A^{-1} \bbm b \) is a solution to \( A \bbm x = \bbm b \).

To show that this solution is unique, suppose that \( \bbm u \) is any solution of \( A\bbm x = \bbm b\), which means that \( A\bbm u = \bbm b\). Multiplying both sides of this equation by \( A^{-1} \) gives \( A^{-1}(A\bbm u) = A^{-1}\bbm b \). Since \( A^{-1}(A\bbm u) = (A^{-1} A)\bbm u = \bbm u \), this shows that \( \bbm u \) must equal \( A^{-1} \bbm b \). \( \Box \)

Notice that this proof required both \( AA^{-1} = I \) and \( A^{-1}A = I \), demonstrating the importance of our inverses being "two-sided."

Properties of Invertible Matrices

We have the following algebraic properties:

You might be thinking that \( (AB)^{-1} \) should equal \( A^{-1} B^{-1} \), but this does not work in general. Why not? When we multiply \( (AB)(A^{-1}B^{-1}) \), we can't "cancel" the \( A \) and \( A^{-1} \) because there is a \( B \) in the way. We can't rearrange the multiplication because matrix multiplication is not commutative. However, \( (AB)(B^{-1}A^{-1}) = A(BB^{-1})A^{-1} = AIA^{-1} = AA^{-1} = I \), without having to change the order of any multiplication. This is why \( (AB)^{-1} = B^{-1}A^{-1} \).

« Lecture 22 Back to Top Lecture 24 »