REINFORCE algorithm



Hi,
Does anyone have an implementation of REINFORCE algorithm by Williams?
My code doesn't converge at all.
Anyway, why doesn't it have much popularity?

Thanks,
Belera Borna

[ comp.ai is moderated. To submit, just post and be patient, or if ]
[ that fails mail your article to <comp-ai@xxxxxxxxxxxxxxxxxx>, and ]
[ ask your news administrator to fix the problems with your system. ]
.



Relevant Pages

  • Re: Tookie Williams Execution - Riots likely?
    ... >> Methinks you're overestimating Mr. Williams' popularity. ... Secondly, Williams isn't Rodney King. ... > Riots aren't about an event, they are triggered by an event. ...
    (alt.true-crime)
  • Re: Tookie Williams Execution - Riots likely?
    ... Secondly, Williams isn't Rodney King. ... Riots aren't about an event, they are triggered by an event. ... King wasn't that popular a guy, it was that the LAPD was unpopular. ... The popularity of the death penalty isn't very high in minority ...
    (alt.true-crime)
  • Re: A bug in difflib module? (find_longest_match)
    ... Simply disabling that ... "popularity check" would slow down the algorithm, ... applied in your case because it forces the algorithm to yield an invalid result. ... and it is also internally used by Differ and others to compare both sequences of lines *and* pairs of similar lines. ...
    (comp.lang.python)
  • Re: bearing calculation off of your heading
    ... I got the below algorithm from Ask Dr.Math that is based on Ed Williams ... I'm trying to figure out the angle between ... a point and the direction of travel of another point. ...
    (sci.geo.satellite-nav)
  • Re: UPA College championship seedings??????
    ... Williams over Pittsburgh? ... a little research and created an algorithm that takes the teams and ... sorts them aesthetically. ...
    (rec.sport.disc)