Author: Rémi Coulom
Date: 13:51:16 06/12/03
Go up one level in this thread
Many people are trying hard to bring new ideas, but the current level of play is so high that it is too difficult to make significant contributions. Most of improvements seem to be highly problem-specific, or even program-specific hand-tuned heuristics. Maybe some have really original ideas, but they are kept secret. Probably the most significant contribution in the last year or so is the web page that Ed wrote when he decided to retire. It proposes a lot of very interesting ideas, an original design philosophy, and some clever tricks. But this looks more like art than science. I tried hard myself to test all kinds of weird ideas, but did not manage to come up with anything really valuable. If I were to do research on computer-chess programming, I would probably concentrate my efforts on learning. Yngvi Björnsson ran some very interesting experiments to learn efficient selectivity heuristics. That looks like a promising research direction. What about you Omid? Are you making plans for a PhD thesis on computer chess? That would sound like a terribly difficult undertaking. You look extremely motivated by computer chess research. I hope you bring us a lot of new ideas. By the way, did you submit a paper to the Graz conference? I hope to meet you there. Greetings, Rémi
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.