Author: Peter Fendrich
Date: 16:26:35 06/29/04
Go up one level in this thread
On June 29, 2004 at 18:21:42, Dan Andersson wrote: > A tall order with both the GA and the later GP part. The GA is just fun during I'm planning for the big one... >set of terminal symbols is hard. I assume you have looked to variations of pure >GP to increase efficiency. F.ex: The Fair Crossover (or Forth Crossover), Gene >Expression Programming, linear gene representations and controlling code bloat. The 'normal' crossover is not as efficient in GP as in GA. As you say, there are a few crossover variations to select from but I don't have the time test them out. I just have to select one that feels good... Never tried GEP but it's probably faster than GP due its simplified approach. Can it be useful here? > Yngvi Bjornsson published some paper using Temporal Learning of extension >values and Multi Cut Pruning in Chess. It might be a good read. Ok, I have read about Temporal difference learning but not by Bjornsson and I'm not sure how Multi Cut Pruning fit in here. > I wish you Good luck, and ungodly processing power. Thanks, Will sure need both luck and cpu power! Mvh Peter >MvH Dan Andersson
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.