Author: blass uri
Date: 22:44:57 07/25/00
Go up one level in this thread
On July 25, 2000 at 18:58:45, Pete R. wrote: >On July 25, 2000 at 14:12:46, Tom Kerrigan wrote: > >>It would be nice to make a change to your evaluation function and get immediate, >>accurate feedback. >> >>So my idea is to get a huge collection of positions of known value (i.e., "white >>has a stronger position") and then see how well the known values correlate to >>the evaluation function's scores. >> >>Does anybody have any ideas for getting a high-quality collection of such >>positions? Or any comments on this approach in general? >> >>-Tom > >I've thought about something similar, i.e. what if God handed you the perfect >eval for a number of positions, that would be wonderful. Obviously if you could >have an eval function that produced the same score that GMs would give a >position each and every time, your program would be unbeatable. I disagree there are cases when GM's are wrong in evaluating. I saw games when GM's over estimate one side's chances and did a mistake because of it. The real reason that GM's are sometimes better than computers is not a better static evaluation function but the fact that they can learn from their search to change their evaluation. GM's look for a plan and do not evaluate positions in the same way as computers. I think that if you give GM's many position and ask them to evaluate what is the expected result for white after 1 second and you give computers the same positions and ask them to evaluate the expected result for white by translating the number of their static evaluation function to the expected result then you are going to find that the evaluation of computers is better because humans cannnot calculate in one second all the things that computers calculate in one evaluation. Uri
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.