Verifying Solutions & Identities - Linear Algebra

  1. For A=[[3,1],[5,2]]A = [[3, 1], [5, 2]] and B=[[2,−1],[−5,3]]B = [[2, -1], [-5, 3]], what is ABAB, and what does it verify?
    AB=IAB = I, since the entries are 6−5=16 - 5 = 1, −3+3=0-3 + 3 = 0, 10−10=010 - 10 = 0, −5+6=1-5 + 6 = 1. So B=A−1B = A^{-1}.
  2. To verify that B=A−1B = A^{-1} for square matrices AA and BB, what single product is enough to check?
    AB=IAB = I (or equivalently BA=IBA = I). For square matrices, a one-sided inverse is automatically two-sided.
  3. For A=[[2,1],[1,2]]A = [[2, 1], [1, 2]] and v=(1,1)v = (1, 1), what scalar λ\lambda satisfies Av=λvAv = \lambda v?
    λ=3\lambda = 3. Av=(2+1,1+2)=(3,3)=3vAv = (2 + 1, 1 + 2) = (3, 3) = 3v, so vv is verified as an eigenvector with eigenvalue 33.
  4. Which equation of the system x1+x2=3x_1 + x_2 = 3, 2x1−x2=12x_1 - x_2 = 1 does the point (x1,x2)=(1,2)(x_1, x_2) = (1, 2) fail to satisfy?
    The second: 2(1)−2=0≠12(1) - 2 = 0 \ne 1. The first holds because 1+2=31 + 2 = 3.
  5. What does one counterexample show about a proposed matrix identity, compared with one example where it holds?
    A single counterexample disproves the identity. An example where it holds proves nothing in general.
  6. How do you verify that a given vector xx is a solution of Ax=bAx = b?
    Compute AxAx and check that it equals bb entry by entry. No row reduction is needed to check a proposed solution.
  7. Why does showing det⁡(A2−3A+2I)=0\det(A^2 - 3A + 2I) = 0 fail to verify that A2−3A+2I=0A^2 - 3A + 2I = 0?
    A nonzero matrix can have determinant 00. A zero determinant only shows the matrix is singular, not that every entry is 00.
  8. For P=[[1,0],[0,0]]P = [[1, 0], [0, 0]], what is P2P^2, and what property of PP does this verify?
    P2=[[1,0],[0,0]]=PP^2 = [[1, 0], [0, 0]] = P, so PP is idempotent (P2=PP^2 = P), as every projection matrix is.
  9. For square matrices, (A+B)2=A2+2AB+B2(A + B)^2 = A^2 + 2AB + B^2 holds exactly when what condition is true?
    AB=BAAB = BA. Expanding gives (A+B)2=A2+AB+BA+B2(A + B)^2 = A^2 + AB + BA + B^2, which matches only if BA=ABBA = AB.
  10. What is the residual of an approximate solution x^\hat{x} to Ax=bAx = b, and what residual means x^\hat{x} is exact?
    The residual is r=b−Ax^r = b - A\hat{x}. The solution is exact exactly when r=0r = 0.
  11. To verify that x=p+tvx = p + tv solves Ax=bAx = b for every scalar tt, which two products must you check?
    Ap=bAp = b and Av=0Av = 0. Then A(p+tv)=Ap+tAv=bA(p + tv) = Ap + tAv = b for every tt.
  12. For A=[[1,1],[0,2]]A = [[1, 1], [0, 2]], what is A2−3AA^2 - 3A, and what identity does it verify?
    −2I-2I. Since A2=[[1,3],[0,4]]A^2 = [[1, 3], [0, 4]], A2−3A=[[−2,0],[0,−2]]A^2 - 3A = [[-2, 0], [0, -2]], which verifies A2−3A+2I=0A^2 - 3A + 2I = 0.

Study these flashcards