Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Temporal Difference

Author: Bas Hamstra

Date: 08:42:16 01/07/01

Go up one level in this thread


On January 07, 2001 at 05:35:32, Gian-Carlo Pascutto wrote:

>On January 06, 2001 at 17:17:06, Bas Hamstra wrote:
>
>>
>>One additional advantage is that when learning from losses only, you don't have
>>to deal with blunders that distort learning. The computer makes no (tactical)
>>blunders. And a blundering opponent will not win against a computer. This (I
>>suspect) increases the probality that IF it learns something, it will be
>>something useful.
>
>One of the papers I have read (I think the KnightCap one) had
>a comparisaion with SAL, and the authors concluded that only
>learning from losses caused the eval function to grade every
>position as a loss.
>
>You're not seeing that effect?

No, not at all. But I remember that they noted that if you lose ALL the games
learning performs bad, you don't refer to that?

Bas.











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.