Computer Chess Club Archives




Subject: Re: Opponent specific learning...

Author: Edward Screven

Date: 14:36:03 03/30/98

Go up one level in this thread

On March 27, 1998 at 14:33:41, Bruce Moreland wrote:

>>On March 26, 1998 at 13:04:34, Dan Homan wrote:
>>>wouldn't be very hard to create a 'profile' for every player
>>>who plays my program and store information about the effectiveness
>>>of certain openings (and other things, as I think of them).

>The programs that play on the chess servers or the web, or commonly play
>in tournaments with multiple opponents, may benefit from being set up to
>do this.  Or perhaps not, since your opponents tend to learn from each
>other's play against you.  In some cases you really are playing against
>one entity -- the user community, which manifests itself in a variety of
>forms, each with individual quirks, and each with a degree of ability to
>learn from the other parts of the community.

if you are trying to play better against a community by learning
from games played against members of the community, then i think
it would pay to have some kind of per-member model.

at a minimum, i think you would want to know the overall strength
of each member.  if your program wins against a weak player, that's
not very interesting, and neither is losing to a very strong player.
in both cases, whatever experience you glean from such games should
have less weight than extracts from games lossed to weak players or
won against strong players.

a half-assed way to do this on a server would be to only learn from
losses, which makes since if you think your program is stronger than
most opponents.

This page took 0.09 seconds to execute

Last modified: Thu, 07 Jul 11 08:48:38 -0700

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