Seonglae ChoReduce matrix size when possible, otherwise, just transpose it
Matrix Derivative Usages
Full cheat sheet
atmos.washington.edu
https://atmos.washington.edu/~dennis/MatrixCalculus.pdf
Core cheat sheet
www.gatsby.ucl.ac.uk
http://www.gatsby.ucl.ac.uk/teaching/courses/sntn/sntn-2017/resources/Matrix_derivatives_cribsheet.pdf