Author: Jay Scott
Date: 09:36:12 08/12/99
Go up one level in this thread
On August 11, 1999 at 17:51:33, H Fraser wrote: >Second, and more to the point, my reading has lead me to modern logic, from >Frege on. What little I've read of computer/AI chess fixes it firmly in >deductive logic and von Neumann machines. Is there a starting point to find out >about any work done/being done in the area of inductive logic, specifically in >ampliative induction? It certainly seems a worthy direction for thinking about >machine chess, especially when skeptical about the possibility of intuition >(machine or human) and facing the dimensional crisis of game theoretic/dynamical >programming approaches. You're gonna scare people off, throwing around old-fashioned terms like "ampliative induction". :-) But seriously, purely deductive approaches are a minority strain in machine learning nowadays. All this neural network and evolutionary algorithm stuff that's flying around falls under the heading of ampliative induction. Of course you wouldn't normally call it logic. For general machine learning in games, I have to recommend my own web site, underhandedly named Machine Learning in Games: http://forum.swarthmore.edu/~jay/learn-game/ If you're interested in inductive logic programming specifically, here are a few things I looked up. This should be enough to get anyone thoroughly bewildered. :-) http://www.cs.bris.ac.uk/~ILPnet2/ http://www.cs.york.ac.uk/mlg/ http://www.cs.utexas.edu/users/ml/ilp.html Jay
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.