Re: [Computer-go] Computer-go Digest, Vol 13, Issue 7

Hi Oliver, unfortunately it's not easy to understand the effects of playout policy modifications on the behavior of MCTS. Some variations indeed refresh so rarely that their suggestions are pretty old most of the time. However, old doesn't necessarily mean outdated, as some types of informatio

Re: [Computer-go] naive thoughts on enthalpy and dynamic komi

IIRC, Hendrik Baier's master thesis shows an experiment conducted using Orego that establishes that Dynamic Komi improved Orego's results against GnuGo. These are non-handicap games. I recall that others have reported the same thing (that dynamic komi helps even in non-handicap games). IMO, Nick

Re: [Computer-go] Game databases

Thanks for sharing your collection Jacques! Very useful for me and others :-) ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go

Re: [Computer-go] Game databases

On Thu, Feb 3, 2011 at 12:30 AM, Adrian Petrescu wrote: > If high-dan amateur games meet your definition of "high-quality", then > uliGo has a large collection of games > from KGS where both players are at least 6d. If you put all the collections > on that page

Re: [Computer-go] naive thoughts on enthalpy and dynamic komi

Shannon's entropy can also be called information. That translates your argument to "you collect more information when the win rate is closer to 50%". Such arguments can be used in support of both dynamic Komi and simulation balancing. I'm not sure how far the argument can be taken, but that's wh

Re: [Computer-go] naive thoughts on enthalpy and dynamic komi

Am 04.02.2011 14:36, schrieb Robert Lupton the Good: Actually that's the Entropy, S; Enthalpy is E - PV (energy corrected for pressure work). If you are interested in thermodynamic analogues, the Helmholtz free energy (E - TS) might be an interesting concept if you can assign meanings to temp

Re: [Computer-go] naive thoughts on enthalpy and dynamic komi

Actually that's the Entropy, S; Enthalpy is E - PV (energy corrected for pressure work). If you are interested in thermodynamic analogues, the Helmholtz free energy (E - TS) might be an interesting concept if you can assign meanings to temperature and energy, as well as Shannon's Entropy.

Re: [Computer-go] naive thoughts on enthalpy and dynamic komi

The crux is "it is gathering it more than twice as fast, which should more than compensate", which is debatable. Infinitely more information may not be enough if you are measuring the wrong thing. On Fri, Feb 4, 2011 at 8:02 AM, Nick Wedd wrote: > In a Monte-Carlo program, the amount of informa

[Computer-go] naive thoughts on enthalpy and dynamic komi

In a Monte-Carlo program, the amount of information derived from one playout is given by its enthalpy -p(win).log_2(p(win)) - p(lose).log_2(p(lose)). This has a maximum at p(win) = 0.5, and is 0 if p(win) is 0 or 1. So, suppose your MC Go-playing program is doing its playouts, and has found s