[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Bug-gnubg] Predefined move filters

From: Gary Wong
Subject: Re: [Bug-gnubg] Predefined move filters
Date: Tue, 31 Dec 2002 14:28:09 -0500
User-agent: Mutt/1.3.28i

On Wed, Jan 01, 2003 at 08:18:46AM +1300, Joseph Heled wrote:
> I cant see how such a tiny number (1296) can give any indication, unless 
> the difference is huge, which I don't think is the case,

I agree; I just suggested 1296 trials to start with for a sanity check.

> and of course 
> you can't use any form of variance reduction, since those depend again 
> on which ply you believe in.

The ply at which variance reduction is performed will affect only the
rate of variance reduction, not the end result.  Variance reduction at
any ply is very helpful, and depending on the other settings, generally
saves at least an order of magnitude of trials.

> For 2 close players (like for a net and a contender) I see I need at 
> least 50,000 money games to give a consistent answer - and that ONLY 
> WHEN I run with "run games in pairs, alternating the the same dice 
> sequence" option, which reduces a lot of the dice noise.

Yes, I think this is helpful (which is why I suggested setting the
seed, so the dice will be matched).

   Gary Wong           address@hidden           http://www.cs.arizona.edu/~gary/

reply via email to

[Prev in Thread] Current Thread [Next in Thread]