Experimenting with policy gradient methods in Jaxgithub.com2 points·by monadicmonad·last year·0 comments
monadicmonad·5 months ago·discussits because benchmarks are nonsense and when you actually use OI models for programming they suck
monadicmonad·last year·discussI don't know how to go from understanding this material to having a job in the field. Just stuck as a SWE for now.