Re:Why are deep learning technologists so overconfident?
by zmgsabst on Wednesday August 31, 2022 @02:22PM
(#32666265)
Deep learning is statistics weighted graphs, but graphs are a way to encode all of mathematics.
Which is why things like manifolds and topology are useful for analyzing our models, because all of the ways to encode mathematics are equivalent and you can transport ideas between them.
Which is why things like manifolds and topology are useful for analyzing our models, because all of the ways to encode mathematics are equivalent and you can transport ideas between them.