Author: Walter Faxon
Date: 22:28:30 07/29/03
Go up one level in this thread
On July 28, 2003 at 04:21:41, Graham Laight wrote: >On July 27, 2003 at 16:29:21, Ricardo Gibert wrote: [snipped] >>L1 L1 R0 L0 L1 L1 R0 R1 L1 R1 L1 L1 R1 L0 R0 L1 L0 L1 L1 R1 L1 R0 R1 R1 L0 L0 L1 >>L1 R0 R1 R0 R0 L0 R1 R1 L0 R0 L0 L1 L1 L1 L0 L0 L0 R1 L0 L0 R1 R0 > >There you go - you won 17 points in the 1st half of the game, but only 10 points >in the 2nd! > >-g Graham, a suggestion: If learning is the key, why not a game that really makes learning pay off? The simplest way in "matching pennies" is to increase the value of each point as the game continues. That is, you incrementally increase the penalty for "exploration" vs. "exploitation". This can lead to meta-strategies where for example you might at first deliberately play predictably (i.e., badly) in order to punish programs that rigidly follow past statistics. I don't know what weight change formula might be best. Maybe several might be tried. The "ultimate" of course is that the weight doubles with each guess. Or instead add a "doubling cube" a la backgammon. Wanna try double or nothing? -- Walter
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.