Author: Dann Corbit
Date: 17:13:20 06/17/02
Go up one level in this thread
On June 17, 2002 at 20:01:39, Russell Reagan wrote: >I'd like to know how many people have to add in a lot of special case code to >their evaluation function(s). For example, does your king safety code in your >evaluation function make your program want to castle, or do you have to add in >code that penalizes the program if it hasn't castled? I suppose I'm wanting to >know how many people have a "natural" evaluation function that takes care of >these kinds of special cases without those cases having to have an extra "work >around" to work. > >Secondly, do you think either approach is superior to the other? I think the >"natural" method would seem to be better. For example, if you reach an endgame >and castling still happened to be legal, and your program castles to get rid of >the "non-castling" penalty, even though your king should be trying to get >centrally located in the middle of the board. If your program had "king safety" >factors instead of the castling work around, then it wouldn't castle and would >play a more correct move. Seems pretty straight forward to me, but I've been >wrong before. > >Other examples could be using a piece-square table with low values for the >border squares and high values for the central squares so that the program >"controls the center" in the opening. Or penalizing pieces that haven't moved >yet in the opening, or penalizing pieces that have moved more than once, etc. >These seem to be very poor methods of approaching an evaluation function at >first thought, but I haven't worked extensively on my own yet, so I'm no expert >here. Castling is a function of game stage. You should castle neither too early nor too late. If you have a function called "game stage" it can possibly tell you if there is a bonus or penaly for castling. Probably most computer programs should be wary of opposite side castling. Pawn storms, and all that. The really good ones have it figured out, I imagine.
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.