Author: Rudolf Huber
Date: 09:11:48 08/02/02
Go up one level in this thread
On August 02, 2002 at 08:24:47, Sune Fischer wrote: >On August 02, 2002 at 02:10:46, Will Singleton wrote: >> >>Fine, though you should know I'm an ex-mac guy. I suggest equal hardware, which >>should be easy to arrange. I don't really care about the tc or venue, whatever >>you want will be okay. Just not bullet. :) >> >>I admire your willingness to take the bet, and I hope you have a huge success. >>But all my experience tells me otherwise. >> >>Will > >I doubt we can conclude a lot about the value of TDLeaf from this competition. >Even if TDLeaf is better than hand tuning, you might still win the match because >other parts of your program is superior. > >I don't think TDLeaf could improve a good hand tuned program by more than 50-100 >points tops, probably less (if any at all of course), so the training method >will be just one small factor in the equation. > >IMHO, if you want to test TDLeaf, you should try it on a program like Crafty :) > >-S. Any respectable result (>=25%) of TDLeaf would be a big surprise to me. In my experience, automatic learning methods quickly learn the simple things but have a hard time with the finer aspects of chess. So for me it is not really necessary to use a common search engine as a base to prove the point that the TDLeaf method has potential. And one should also note that hand tuning will profit from public source code which gives it a big advantage. Rudolf
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.