7.1-exam3-2013¶
Quiz 3, 2013 · Section 7.1: Singular Values and Singular Vectors Source: exam3_2.pdf, pp. 55, 63-64
Problem¶
1. (32 pts.) (2 points each)
There are sixteen \(2 \times 2\) matrices whose entries are either 0 or 1. For each of the sixteen, write down the two singular values. Time saving hint: if you really understand singular values, then there is really no need to compute \(AA^T\) or \(A^TA\), but it is okay if you must.
Solution
Useful facts: * Row/column swaps don't alter singular values. * The sum of squares of the singular values equals the sum of squares of the entries of the matrix. * For a symmetric matrix, the singular values are the absolute values of the eigenvalues. * \(\begin{bmatrix} 0 & 0 \\ 0 & 0 \end{bmatrix}\): singular values \(0, 0\). * \(\begin{bmatrix} 1 & 0 \\ 0 & 0 \end{bmatrix}, \begin{bmatrix} 0 & 1 \\ 0 & 0 \end{bmatrix}, \begin{bmatrix} 0 & 0 \\ 1 & 0 \end{bmatrix}, \begin{bmatrix} 0 & 0 \\ 0 & 1 \end{bmatrix}\): all related to \(\begin{bmatrix} 1 & 0 \\ 0 & 0 \end{bmatrix}\) by row/column swaps. \(\begin{bmatrix} 1 & 0 \\ 0 & 0 \end{bmatrix}\) is symmetric with eigenvalues (1, 0), so its singular values are \(1, 0\). * \(\begin{bmatrix} 1 & 1 \\ 0 & 0 \end{bmatrix}, \begin{bmatrix} 1 & 0 \\ 1 & 0 \end{bmatrix}, \begin{bmatrix} 0 & 1 \\ 0 & 1 \end{bmatrix}, \begin{bmatrix} 0 & 0 \\ 1 & 1 \end{bmatrix}\): all the same as \(\begin{bmatrix} 1 & 1 \\ 0 & 0 \end{bmatrix}\). It has rank 1, so one singular value is 0. If \(\sigma\) is the other, \[ 0^2 + \sigma^2 = 1^2 + 1^2, \] so \(\sigma = \sqrt{2}\). Singular values \(\sqrt{2}, 0\). * \(\begin{bmatrix} 1 & 0 \\ 0 & 1 \end{bmatrix}, \begin{bmatrix} 0 & 1 \\ 1 & 0 \end{bmatrix}\): related by a column swap. \(\begin{bmatrix} 1 & 0 \\ 0 & 1 \end{bmatrix}\) has eigenvalues 1, 1, so singular values \(1, 1\). * \(\begin{bmatrix} 1 & 1 \\ 1 & 0 \end{bmatrix}, \begin{bmatrix} 0 & 1 \\ 1 & 1 \end{bmatrix}, \begin{bmatrix} 1 & 1 \\ 0 & 1 \end{bmatrix}, \begin{bmatrix} 1 & 0 \\ 1 & 1 \end{bmatrix}\): all the same as \(\begin{bmatrix} 1 & 1 \\ 1 & 0 \end{bmatrix}\), which has eigenvalues \[ \frac{1 + \sqrt{5}}{2}, \frac{1 - \sqrt{5}}{2}. \] The second one is negative, and since the matrix is symmetric the singular values are the absolute values of the eigenvalues: \[ \frac{\sqrt{5} + 1}{2}, \frac{\sqrt{5} - 1}{2}. \] * \(\begin{bmatrix} 1 & 1 \\ 1 & 1 \end{bmatrix}\): symmetric with eigenvalues (0, 2), so its singular values are \(0, 2\).Keywords: singular values, singular value decomposition, eigenvalues, symmetric matrix, rank