Author: Steffen Jakob
Date: 08:39:09 08/10/98
Go up one level in this thread
On August 10, 1998 at 10:16:23, Robert Hyatt wrote: >>>I would like to tell you something about an idea which I had this >>>weekend. I talked a bit about it with Bruce at ICC and he said that it >>>might already be a known idea. I would appreciate some feedback. >> >>>Some time ago someone (KK?) posted statistics about the first moves of >>>white. This seemed rather useless for me. More interesting would be >>>the the likelihood of _any_ move played in a won game by a strong >>>player. This information e.g. could be used for a trivial move >>>ordering. My plan is to do the following: >> >>>1. Use a precalculated list of all possible moves for the move >>> generator in your chess engine. >>>2. Add a counter to the move structure >>>3. scan a large database with only high quality games and >>> increment the counter for each game where the winner made this >>> move. >> >>You will find the idea in Rebel10. It will be an option as I don't know >>if it makes the program stronger or weaker. The idea goes further in >>respect that results found in the database are subject to the Rebel score. >I think you misunderstood, based on comments I had with him on ICC this >morning. IE he was talking about using this to *order* moves, not affect >the final score, as I understood him. IE if a GM played this move most, >then order that first wherever the position occurs in the tree... > >Which I don't think will work well at all, since that might take a deep >search before it looks best to a program, and putting a non-best (to the >program) move at the top of the move list will hurt and not help the search >speed... > >Unless I misunderstood everything he said, that is.. :) Actually move ordering was the first thing which came into my mind how I could use the counter data and that was what I told you when we chatted in ICC today. But as I mentioned in another posting it would be also possible to use the counter values for the evaluation itself. I didnt know that Ed is already doing this with Rebel :-). If you doing this, the move ordering makes sense, too, because then move order should correlate with the evaluation function. >>For every move found in the database Rebel will give a flexible bonus or >>penalty based on the percentage and average ELO found in the database. >> >>In this way it is the database who will guide Rebel which move to play. >>Blunders are automatically filtered because the search will decide after >>all. >>>The counter could be used to perform a very simple but also cheap move >>>ordering or even for the evaluation itself. This idea could be improved >>>if you don't store only a counter but a set of (pawnstruct, counter) >>>pairs. Best wishes, Steffen.
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.