Author: Fernando Villegas
Date: 10:08:08 10/19/00
For what we would want self-learning programs? Most of us, poor, sad dudes, are uncapable of learning by ourselves and so are in need of a device for not falling once and again in the same mistakes. And it is so easy that I marvel nobody has implemented yet. This is how it should operate: a) every time in a game we make a move that is an awfull mistake -score goes down- the program should store the position and the move. Next time we fall in the same position and if we make the same bad move, then a dialog should appears with a ramdom collection of insults and deprecations, warning us of our stupidity and asking us to look something better. Similar to the tutor watching function of Fritz, but with the difference he would recall us that we already did the mistake before. So simple like that but more efective. b) How to do it? Just a little database of wrong moves and its asociated positions and a little add-on or sub-program in the code checking every position and move to see if it fits to one of the database. Even I could write that...although just in basic. Fernando
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.