[peytn]

I dunno, there are definitely distribution-based assumptions—good luck working with skewed data. Most old-school techniques are kinda additive, so nobody's really been assuming a single distribution for practical applications.

Current ML techniques just work well for the kinds of problems people are applying them to, which is kind of a tautology. We should definitely seek to understand the theory behind stuff like dropout and not consider our lack of understanding a strength.