Skip to content

Comment on The "Soft Maximum" functionparent

Comments

True. To keep the explanation simple I glossed over some details. The problem you describe exists even in the 1D case, e.g. the function f(x) = 0. However if a convex function has multiple minima then the minima will be connected and at the same height, so the algorithms still apply.

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.