In Reply to: Lecture Notes Set #7 p.10 posted by Doug on November 24, 2002 at 11:41:48:
>Hi Professor Tao,
>In the proof, you said that "Write D = diag (lambda_1,...,lambda_n, so that De_j = lambda_j". However, in the subsequent line, there is an intermediate step which says "Q*D*e_j = Q*lambda_j*e_j".
>I doubt why the e_j remains in the derivation?
>Thanks!
Oops! The equation "D e_j = lambda_j" should read
"D e_j = lambda_j e_j" instead. (D e_j = lambda_j doesn't
make sense, since the left-hand side is a vector and the
right-hand side is a scalar).
I've corrected the Week 6 and 7 notes accordingly.
Terry