Confusion about finding V given U and S during SVD

45 Views Asked by At

I understand any rectangular mxn matrix A is a factorization of the form: $$ A = U\Sigma V^{\top}$$ where U and V are orthogonal matrices and $\Sigma$ is diagonal.
Say, I have found U and $\Sigma$ and looking for V. $$ A = U\Sigma V^{\top} \\ AV = U\Sigma \\ Av_i = \sigma_i u_i \\ A^{\top}Av_i = A^{\top}\sigma_i u_i \\ v_i = A^{\top}\sigma_i u_i \\ v_i = \sigma_iA^{\top}u_i \\ $$ Also from my notes I have, $$ A = U\Sigma V^{\top} \\ A^{\top}A = A^{\top}U\Sigma V^{\top} \\ I = A^{\top}U\Sigma V^{\top} \\ A^{\top}U = V\Sigma^{\top} \\ (A^{\top}U)_i = v_i\sigma_i \\ v_i = \frac{(A^{\top}U)_i}{\sigma_i} $$

What am I doing wrong to get these conflicting equations?