4.5 Exemples de matrices élémentaires

Soit \(\mathbf{A} = \begin{bmatrix} 1 & 0 & 2\\ -2 & 0 & -3 \\ 0 & 2 & 0 \end{bmatrix}.\) Nous allons obtenir la FER de \(\mathbf{A}\) en utilisant les opérations élémentaires sur les lignes. Dans ce qui suit, chaque étape montre l’état actuel de la matrice et l’opération élémentaire sur les lignes à appliquer afin d’obtenir la prochaine matrice. La matrice élémentaire correspondant à chaque opération est indiquée en parenthèses.

\[\begin{align*} & \begin{bmatrix} 1 & 0 & 2 \\ -2 & 0 & -3 \\ 0 & 2 & 0 \end{bmatrix} \\ \xrightarrow{L_2 \leftarrow L_2 + 2 L_1} & \begin{bmatrix} 1 & 0 & 2 \\ 0 & 0 & 1 \\ 0 & 2 & 0 \end{bmatrix} && \left(\mathbf{M}^{(1)} = \begin{bmatrix}1 & 0 & 0\\ 2 & 1 & 0\\ 0 & 0 & 1\end{bmatrix}\right) \\ \xrightarrow{L_2 \leftrightarrow L_3} & \begin{bmatrix} 1 & 0 & 2 \\ 0 & 2 & 0 \\ 0 & 0 & 1 \end{bmatrix} && \left(\mathbf{M}^{(2)} = \begin{bmatrix}1 & 0 & 0\\ 0 & 0 & 1\\ 0 & 1 & 0\end{bmatrix}\right) \\ \xrightarrow{L_2 \leftarrow \frac{1}{2}L_2} & \begin{bmatrix} 1 & 0 & 2 \\ 0 & 1 & 0 \\ 0 & 0 & 1 \end{bmatrix} && \left(\mathbf{M}^{(3)} = \begin{bmatrix}1 & 0 & 0\\ 0 & \frac{1}{2} & 0\\ 0 & 0 & 1 \end{bmatrix}\right) \\ \xrightarrow{L_1 \leftarrow L_1 + (-2)L_3} & \begin{bmatrix} 1 & 0 & 0 & 1 \\ 0 & 1 & 0 & -1 \\ 0 & 0 & 1 & -1 \end{bmatrix} && \left(\mathbf{M}^{(4)} = \begin{bmatrix}1 & 0 & -2\\ 0 & 1 & 0\\ 0 & 0 & 1\end{bmatrix}\right) \end{align*}\]

La dérivation précédente montre que \[ \begin{bmatrix} 1 & 0 & 2 \\ 0 & 0 & 1 \\ 0 & 2 & 0 \\ \end{bmatrix} = \mathbf{M}^{(1)}\mathbf{A}, \] \[ \begin{bmatrix} 1 & 0 & 2 \\ 0 & 2 & 0 \\ 0 & 0 & 1 \end{bmatrix} = \mathbf{M}^{(2)}\begin{bmatrix} 1 & 0 & 2 \\ 0 & 0 & 1 \\ 0 & 2 & 0 \\ \end{bmatrix} = \mathbf{M}^{(2)}(\mathbf{M}^{(1)}\mathbf{A}), \] \[ \begin{bmatrix} 1 & 0 & 2 \\ 0 & 1 & 0 \\ 0 & 0 & 1 \end{bmatrix} = \mathbf{M}^{(3)}\begin{bmatrix} 1 & 0 & 2 \\ 0 & 2 & 0 \\ 0 & 0 & 1 \end{bmatrix} = \mathbf{M}^{(3)}(\mathbf{M}^{(2)}(\mathbf{M}^{(1)}\mathbf{A})), \] \[ \begin{bmatrix} 1 & 0 & 0 \\ 0 & 1 & 0 \\ 0 & 0 & 1 \end{bmatrix} = \mathbf{M}^{(4)}\begin{bmatrix} 1 & 0 & 2 \\ 0 & 1 & 0 \\ 0 & 0 & 1 \end{bmatrix} = \mathbf{M}^{(4)}(\mathbf{M}^{(3)}(\mathbf{M}^{(2)}(\mathbf{M}^{(1)}\mathbf{A}))). \]

D’après le dernière égalité, on remarque que \(\mathbf{M}^{(4)}(\mathbf{M}^{(3)}(\mathbf{M}^{(2)}(\mathbf{M}^{(1)}\mathbf{A}))) = \begin{bmatrix}1&0&0\\0&1&0\\0&0&1\end{bmatrix}\). Le côté gauche est en désordre, mais nous savons maintenant qu’il existe une matrice \(\mathbf{M}\) telle que \[\mathbf{M}\mathbf{A} = \mathbf{M}^{(4)}(\mathbf{M}^{(3)}(\mathbf{M}^{(2)}(\mathbf{M}^{(1)}\mathbf{A}))).\] En effet, il découle du théorème 4.1 que \[\mathbf{M}^{(4)}(\mathbf{M}^{(3)}(\mathbf{M}^{(2)}(\mathbf{M}^{(1)}\mathbf{A}))) = (\mathbf{M}^{(4)}(\mathbf{M}^{(3)}(\mathbf{M}^{(2)}\mathbf{M}^{(1)})))\mathbf{A}.\] Nous pouvons d’abord calculer \(\mathbf{M}^{(2)}\mathbf{M}^{(1)}\), puis calculer \(\mathbf{M}^{(3)}(\mathbf{M}^{(2)}\mathbf{M}^{(1)})\), et finalement calculer \(\mathbf{M}^{(4)}(\mathbf{M}^{(3)}(\mathbf{M}^{(2)}\mathbf{M}^{(1)})).\) Cela nous donne \(\mathbf{M} = \begin{bmatrix} -3 & -2 & 0\\ 0 & 0 & \frac{1}{2}\\2 & 1 & 0 \end{bmatrix}\). Il est facile de vérifier que \(\mathbf{M}\mathbf{A} = \mathbf{I}_3.\)

Remarque. S’il n’est pas nécessaire de spécifier chacune des matrices élémentaires, on peut obtenir \(\mathbf{M}\) directement en appliquant la même série d’opérations élémentaires sur les lignes à la matrice identité \(\mathbf{I}_3\). (À essayer.)

La matrice \(\mathbf{M}\) est l’inverse à gauche de \(\mathbf{A}.\) Lorsque l’on la multiplie à gauche de \(\mathbf{A}\), on obtient la matrice identité. Ce aussi le cas si nous multiplions à la droite de \(\mathbf{A}\), c’est-à-dire que \(\mathbf{A}\mathbf{M} = \mathbf{M}\mathbf{A} = \mathbf{I}_3.\) Comme nous allons le voir dans la section suivante, ce n’est pas un coïncidence.

Exercices

  1. Soit \(\mathbf{A} \in \mathbb{R}^{5 \times m},\) \(m \geq 1.\) Soit \(\mathbf{M} = \begin{bmatrix} 1 & 0 & 0 & 0 & 0 \\ 0 & 0 & 1 & 0 & 0 \\ 0 & 0 & 0 & 0 & 1 \\ 0 & 0 & 0 & 1 & 0 \\ 0 & 1 & 0 & 0 & 0 \end{bmatrix}.\) À quoi équivalant le produit \(\mathbf{M}\mathbf{A}\)?

    1. Échanger les lignes 2 et 3, pour ensuite échanger les lignes 3 et 5.

    2. Échanger les lignes 2 et 3 pour ensuite échanger les lignes 2 et 5.

    3. Échanger les lignes 3 et 5 pour ensuite échanger les lignes 2 et 3.

  2. Trouvez un inverse à gauche de chacune des matrices suivantes.

    1. \(\begin{bmatrix} 3 & 4 \\ 2 & 3 \end{bmatrix}\)

    2. \(\begin{bmatrix} 1 & 0 & 0 \\ 2 & 1 & 0 \\ 3 & 2 & 1\end{bmatrix}\)

Solutions

  1. Échanger les lignes 2 et 3 et ensuite échanger les lignes 3 et 5.

    1. \(\begin{bmatrix} 3 & -4 \\ -2 & 3 \end{bmatrix}\).

    2. \(\begin{bmatrix} 1 & 0 & 0 \\ -2 & 1 & 0 \\ 1 & -2 & 1 \end{bmatrix}\).