Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Anyone for an 'increasing expectation' heuristic?

Author: Francis Monkman

Date: 02:04:47 05/03/99

Go up one level in this thread



On May 02, 1999 at 19:25:01, Will Singleton wrote:

>I think a well-used method for determining the order of root moves is to see how
>long each takes, then sort (excluding the best move) based on longest time
>taken.  So, I suppose you could look at the root order at each iteration, and
>see if any moves have moved significantly up the list.  But I don't know about
>extending on such information; in fact, I don't think that info is very
>significant, other than for root ordering purposes.
>
>Or do you have another method for determining epectations?
>
>Will

The idea is more or less along these lines: keep the score for each move at each
ply-depth of search, then the resulting 'statistical curve(s)' can be used to
determine whether a move is gaining in 'expectation-value' (and hence should
definitely be explored thoroughly). Conversly, programs often seem to choose
moves that seem stronger,even though they are losing expectation-value as the
search deepens -- perhaps this might be avoided by such a  method, too.

Francis



This page took 0 seconds to execute

Last modified: Thu, 15 Apr 21 08:11:13 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.