Author: Uri Blass
Date: 20:21:00 09/19/00
Go up one level in this thread
On September 19, 2000 at 22:19:12, pavel wrote: >Hi, > I have been thinking about a learning idea for chess programs for some time, >and brought this idea once before in this forum, but probably didnt get enough >feedback, or explanation werent clear enough. > here is what I wanted to know.... > >lets make a chess program (program X) analyze a set of games which contains 3000 >positions, program X will analyze for each position for,lets say, 10secs; other >then the moves that are already in the book. > >It will store the evaluation value, along with the 'depth' and 'next best move' >it thought of, for each position in a *bat or *txt file, >Later, when it comes to that position again, it will start calculating from >where it left, eg, if it left in depth 6 in e2e4 it will start analyzing from >then on, since the depth and the pV will be stored. Depth and pv is not enough to continue from the same point that it left because programs remember a lot of things in the has tables. Another point is that programs have no idea about the next best move >It will make it search faster, or(and) avoid searching bad positions. I am not sure about it. I also think that most of the positions will not be relevant in games. A better idea seems to be to give the program a very big database of games and tell it to analyze the most common positions that are not in the book and are relevant for it(it should find the most common positions from all the possible positions that it can find itself out of book) The positions that it can find itself out of book are all the positions when the book ends and all the positions that are 1 ply after a book move when there is no response in the book. It can always search for the most common position in the big database and analyze it. The program can analyze these positions for some minutes and add the result of the search to the book so when it plays in a game it can save time in these positions. It should also remember the time it searched and the number of nodes per second so in the cases that it plays in time control that is slow enough it is not going to trust the book moves that it generated. 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.