Author: Robert Hyatt
Date: 06:38:30 09/28/99
Go up one level in this thread
On September 28, 1999 at 05:23:35, Gunnar Andersson wrote: >On September 27, 1999 at 13:59:58, Robert Hyatt wrote: > >>Picking good book moves is an issue. The commercial guys seem to bank on >>having a human tune their book. This is impractical, for lots of reasons. >>Letting the program tune its own book is also very difficult. There really >>isn't a good solution for this problem yet, either human solution or computer >>solution... > >Isn't this because the evaluation functions haven't reached the level where the >values they return have a good global interpretation? I.e. comparable between >different game stages and different types of games. I don't think this is the issue. Here are some problems that have to be overcome: 1. no "evaluation" can pick a good book move without a deep search to go with it. Why? Because humans can't pick good book moves in most positions without a lot of prior searching (done by others) to guide them. Hence the plethora of books like MCO, ECO, BCO, etc... 2. frequency of play is not a good guideline in many cases, because a recent refutation gets swamped by all the times it was played successfully in the past. 3. opening theory has many outright mistakes that are waiting to get found. At the highest levels of play, opening preparation is hugely important. As of now, computers are not good at doing this. > >I've been involved in computer Othello for a while now. It's considerably easier >to tune a good evaluation function automatically and all the top programs (which >are much stronger than the human world champion) tune their book automatically. >In the future, evaluation functions in chess programs might reach this level too >and then having humans tune the book won't be necessary (or a good idea at all). > >/ Gunnar Chess is a much more complex game, unfortunately. I don't think it will ever be possible to have an evaluation function that will lead to GM-quality play, without an _extremely_ deep search to go along with it. This isn't necessarily true in othello, and the game is a lot less 'wide' as well, tree-wise...
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.