News for hackers. Stuff that matters.

A comment on Machine Learning from scratch: Bare bones implementations in Python. Parent comment

Re:Machine Learning from scratch: Bare bones implementations... by JustFinishedBSG on Saturday February 25, 2017 @03:22PM (#13733627)
For solving linear systems there are recent mind-blowing methods by R. M. Gower and P. Richtarik:

https://arxiv.org/abs/1506.03296

http://www.maths.ed.ac.uk/~richtarik/papers/SDA.pdf

Plus R. M. Gower is fantastically nice and enthusiastic so there's that

And thanks for the link !

[ Parent | on HN ]
Re:Machine Learning from scratch: Bare bones implementations... by stabbles on Sunday February 26, 2017 @04:41AM (#13736472)
Only scrolled through it, but AFAIK if its results are comparable to Kaczmarz, then note that it is extremely slow.

If your goal is to solve say a LS problem, why not go for CGLS? http://web.stanford.edu/group/SOL/software/cgls/

[ Parent | on HN ]
Re:Machine Learning from scratch: Bare bones implementations... by zzleeper on Saturday February 25, 2017 @11:13PM (#13735633)
Has anyone implemented them in production? I read their paper a while ago, but thought that the dual methods they showed might have practical limitations (like insane cache misses, compared to more standard methods)
[ Parent | on HN ]