Author: Graham Laight
Date: 15:53:24 01/10/99
Go up one level in this thread
On January 09, 1999 at 09:21:41, Laurence Chen wrote: >On January 09, 1999 at 05:55:25, Graham Laight wrote: > >>As I was sitting eating my breakfast just now, it occured to me that there are >>basically 3 items that, between them, will influence how close an evaluation of >>a chess position is to how good that position really is: >> >>1. The number of pieces of knowledge the evaluation function can call upon >> >>2. The quality of those pieces of knowledge >> >>3. The accuracy of selecting the right pieces of knowledge (and their >>appropriate weightings) for the position at hand >> >> >>Does anybody have any thoughts about this? >I thought about this before, and I just wanted to ask you how would you >determine that the chess engine is assessing the position accurate? Unless one >has a good understanding of the position oneself, how can one judge that the PV >evaluation of the engine is correct? Sometimes, even I make such a bad mistake >of evaluating and judging a chess position, I would think that White had a >better position, and the chess engine believes that it is Black who has the >better position, and to complicate matters, if one has several different >engines, one can observe that the engines don't all agree, one engine will think >that White has the better position and the other one Black has the better >position. So how do you know that the chess engine evaluation of the position is >correct then? >Regards, >Laurence The ultimate answer would, of course, be perfect knowledge of whether a particular position would, with best play, result in a win, a draw or a loss. In reality we do not know this for most positions. In which case there are 2 measures: 1. Assessment by giving the system a large number of games and scoring the results - which is what the SSDF do. 2. In the absence of a large number of games, like Deeper Blue, one has to ask experts to assess the moves that the system made. The consensus of the experts who analysed DB's moves against Gary Kasparov was that in played some really excellent chess at times, but played some lousy moves at other times (by GM standards).
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.