News for hackers. Stuff that matters.

A comment on Machine learning works spectacularly well, but mathematicians aren’t sure why. Parent comment

Re:Machine learning works spectacularly well, but mathematic... by eli_gottlieb on Friday December 04, 2015 @10:52AM (#10676962)
>The very notion of generalization to unobserved data is not well understood (I like D. Wolpert papers on that topic).

Which papers?

[ Parent | on HN ]
Re:Machine learning works spectacularly well, but mathematic... by cdavid on Friday December 04, 2015 @11:42AM (#10677317)
IIRC, that one was a decent overview, though he has worked on similar issues in older papers: http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.99.1...

It has been a while since I read those papers, there may better references.

[ Parent | on HN ]