You Don't Need Matrix Calculus

Matrix calculus, popular in machine learning, offers rules for differentiating vector and matrix functions but requires memorizing complex rules that depend on row/column conventions. The author argues you can skip it entirely: just write out the indices and use ordinary differentiation. This always works, needs no new rules, and is often faster. A neuron activation example shows deriving three derivatives at once, contrasting with a convoluted matrix calculus result.