Author: Don Dailey
Date: 18:20:29 01/28/98
Go up one level in this thread
On January 28, 1998 at 17:36:06, Bruce Moreland wrote: > >On January 28, 1998 at 16:38:14, Don Dailey wrote: > >>>I am not a big Carl Sagan fan, but in one of his books, I believe it was >>>The Dragons of Eden, I think he suggested that intelligence is an >>>individual's capacity for adaptation. Sorry for the vagueness but I >>>read all of his stuff in approximately 1980. >>> >>>Personally, I think this definition is a good one, better than the one >>>in the OED at describing this part of the word's meaning. >> >>This is not a bad definition. There is a real sense that most chess >>programs do not adapt (but there programmers do from version to >>version.) >> >>The learning stuff in some of the programs does try to provide >>adaptation >>at least weakly. > >I think they adapt like crazy. Put some pieces on the board and see >what happens. They'll always try to find a good move, and often they >will succeed, they are general-purpose problem solvers, within the >domain of chess. Yes I guess they do adapt. I was thinking about it from the learning point of view and didn't consider the adaptations they constantly perform. >I don't think that learning is a requirement for an AI topic, is it? It might be actually. I haven't heard a formal definition in a long time but I do remember memory being considered a very important part of intelligence. You could probably easily argue that this requirement is met with chess programs. I may have assumed the memory part had to do with learning. You could even argue that the search is learning too (boy this is a slippery subject.) Remember though, I'm one of those that believe it's proper to consider chess programs as AI. >Adaptation doesn't have to involve repeated encounters with a similar >situation, it can come into play when you see a completely new >situation. > >There are lots of places to apply what has traditionally been thought of >as AI to chess, not necessarily just in the search. I think Crafty's >learning is an interesting AI experiment -- Bob is teaching Crafty to >play a series of chess games well rather than just one. > >I think he should find some examples where it has adapted in order to >improve its results against specific opponents, and write something up, >I think it would make an interesting article. > >bruce
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.