Author: Christophe Theron
Date: 14:48:44 10/21/99
Go up one level in this thread
On October 21, 1999 at 14:30:46, Dave Gomboc wrote: >On October 21, 1999 at 12:09:18, Christophe Theron wrote: > >>On October 21, 1999 at 06:29:12, Dave Gomboc wrote: >> >>>On October 20, 1999 at 09:41:21, Christophe Theron wrote: >>> >>>>BTW: Tiger has NO agressive learning. It does not try to repeat won games. It >>>>just try to avoid repeating losses. This game appears twice only by bad luck... >>>> >>>> >>>> Christophe >>> >>>In auto-232 play, isn't this a mistake? >>> >>>Dave >> >>Why? Because I could get easily some extra points by beating non-learning >>programs several times? > >No... because other programs will be attacking yours in that way, and if you're >otherwise equally strong, you'll score less than 50%. > >Of course, if you can spend the time that you saved not working on book learning >much to ensure the programs are not going to be "otherwise equal", that's fine >too. :-) I'll use agressive learning when I'll not be able to improve the engine. But I have a todo list that is so long that I can't believe this will ever happen! :) >>I don't consider book learning to be very smart. So I have done the smallest >>book learning I could. I don't want my program to be called stupid so I don't >>repeat lost games. >> >>I think customers will appreciate this. Tiger assumes they are not stupid and >>won't play a game they have lost already. So what's the point in trying to do >>it? >> >>But Tiger knows customers are nasty sometimes and will certainly try to beat him >>twice in the same way! :) >> >>BTW I think that in Auto232 the real mistake is to waste time trying to get >>points unfairly by writting an agressive book learner. This time would be better >>used in improving the engine. > >Quite possibly. > >> Christophe > >Dave
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.