Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

having to do with distributed markov chains

I'm guessing you mean distributed Markov Chain Monte Carlo ?

Sounds like an interesting problem, I'd really like to hear more about that.

I'm really interested in "difficult" problems with significant commercial applications. (My contact info is in my profile if you don't want to post publicly).



Parallel MCMC implementations exist. My favorite is emcee - http://dan.iel.fm/emcee/current/ which has excellent support for parallelism.

Derail aside - I completely agree that lying about features is horrendous and I'd never ever work for a boss that did this on a regular basis.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: