18.065 LEC4. Eigenvalues and Eigenvectors

《Matrix Methods in Data Analysis, Signal Processing, and Machine Learning》MIT Course Number：18.065课程习题答案

LEC4. Eigenvalues and Eigenvectors

Problems of Lecture 4 (from textbook Section I.6)

2 Compute the eigenvalues and eigenvectors of A and $A^{-1}$. Check the trace !
$$A=\left[\begin{array}{cc}0& 2\\ 1& 1\end{array}\right]and{A}^{-1}=\left[\begin{array}{cc}-1/2& 1\\ 1/2& 0\end{array}\right].$$$A^{-1}$ has the $\underline{\ast \ast \ast \ast }$eigenvectors as A. When A has eigenvalues ${\lambda }_1$ and ${\lambda }_2$ , its inverse has eigenvalues $\underline{\ast \ast \ast \ast }$.

11 The eigenvalues of A equal the eigenvalues of $A^T$. This is because det(A − λI) equals det($A^T$ − λI). That is true because $\underline{\ast \ast \ast \ast }$ . Show by an example that the eigenvectors of A and $A^T$ are not the same.

15 (a) Factor these two matrices into A = $X\Lambda$$X^{-1}$ :
$$A=\left[\begin{array}{cc}1& 2\\ 0& 3\end{array}\right]andA=\left[\begin{array}{cc}1& 1\\ 3& 3\end{array}\right].$$
(b) If A = $X\Lambda$$X^{-1}$ then $A^3$ = ( )( )( ) and $A^{-1}$ = ( )( )( ).

Solutions to Lecture 4

2 $A^{-1}$ has the $\underline{same}$ eigenvectors as A. When A has eigenvalues ${\lambda }_1=2$ and ${\lambda }_2=-1$ , its inverse has eigenvalues $\underline{{\lambda }_1^{-1}=1/2and{\lambda }_2^{-1}=-1}$.

11 (A − λI) has the same determinant as ($A^T$ − λI) which can be written as $(A-\lambda I{)}^T$ because every square matrix has det(M) = det($M^T$) . That every square matrix has det(M) = det($M^T$) is a basic fact because determinant is the sum of product of items that come from different columns and rows and if you transpose that matrix the result won’t change.
$$\left[\begin{array}{cc}1& 0\\ 1& 0\end{array}\right]haveeigenvectors\left[\begin{array}{c}0\\ 1\end{array}\right],\left[\begin{array}{c}1/\sqrt{2}\\ 1/\sqrt{2}\end{array}\right]$$
$$\left[\begin{array}{cc}1& 1\\ 0& 0\end{array}\right]haveeigenvectors\left[\begin{array}{c}-1/\sqrt{2}\\ 1/\sqrt{2}\end{array}\right],\left[\begin{array}{c}1\\ 0\end{array}\right]$$They have different eigenvectors.

15 (a)
$$A=\left[\begin{array}{cc}1& 2\\ 0& 3\end{array}\right]=\left[\begin{array}{cc}1& 1/\sqrt{2}\\ 0& 1/\sqrt{2}\end{array}\right]\left[\begin{array}{cc}1& 0\\ 0& 3\end{array}\right]\left[\begin{array}{cc}1& -1\\ 0& \sqrt{2}\end{array}\right]$$
$$A=\left[\begin{array}{cc}1& 1\\ 3& 3\end{array}\right]=\left[\begin{array}{cc}1/\sqrt{2}& 1/\sqrt{10}\\ -1/\sqrt{2}& 3/\sqrt{10}\end{array}\right]\left[\begin{array}{cc}0& 0\\ 0& 4\end{array}\right]\left[\begin{array}{cc}3\sqrt{2}/4& -\sqrt{2}/4\\ \sqrt{10}/4& \sqrt{10}/4\end{array}\right]$$
(b) If A = $X\Lambda$$X^{-1}$ then $A^3$ = ($X$)(${\Lambda }^3$ )($X^{-1}$) and $A^{-1}$ = ($X^{-1}$)(${\Lambda }^{-1}$)($X$).