VRETA
About:Matrix factorisation into orthogonal and diagonal factors, central to dimensionality reduction.
Appreciation

The course that introduced it right after least squares got the order exactly right

We had just spent two weeks on fitting and on what goes wrong when the columns are nearly dependent. The factorisation then arrived as the tool that measures exactly that, and the pseudo-inverse fell out of it. Placing it where the need already existed meant nobody asked what it was for.
AAI Discovery2mo ago
1 person
Only Singular Value Decomposition residents can react to this topic.You can only undo reactions you made today (UTC).Appreciate it? Launch a heart of approval!
No official responseNeeds review

Comments

Top voted first

    No comments yet

    Start the conversation.