Author: Djordje Vidanovic
Date: 12:26:02 07/14/05
Go up one level in this thread
On July 14, 2005 at 14:55:31, Dann Corbit wrote: >On July 13, 2005 at 23:16:05, Steven Edwards wrote: >[snip] >>So, the cognitive search mimics human search by executing one chunk (not one >>move) of planning/exploring after another, hopping from one part of the tree to >>anther revisiting plans and nodes with increasing effort, progressing in both >>width and depth, reducing global uncertainty, and selecting a top level move in >>the available time. I know of no other two person game program that searches in >>this way. Unless there's already a name for it, I'm going to call it the P >>search (P = Progressive). > >How do you know that this mimics human search? >Sometimes, when I play, I am trying to improve my position and not looking for >tactics {indeed, I do not begin to look for tactics until I have a better >position} >Sometimes, I think about a very long range plan. >Sometimes, I think about a near term tactical goal like a pawn to snack on (bad >habit gathered from too many computer chess games} >Sometimes, I pick out a piece that annoys me and try to build a huge battery >against it. >Sometimes, I aim to pressure the king. > >I don't know how chunks might come into play, and the way I think about what I >am going to do varies quite a bit. > >Of course, you might be modelling how a GOOD player makes his moves. >;-) Steven is actually very close to what's been done. I haven't got my ICCA journals at hand now, but they published a paper that described 'chunking for knowledge'. You might want to take a look. Djordje
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.