KnightCap: A chess program that learns by combining TD(lambda) with game-tree search

by   Jonathan Baxter, et al.

In this paper we present TDLeaf(lambda), a variation on the TD(lambda) algorithm that enables it to be used in conjunction with game-tree search. We present some experiments in which our chess program "KnightCap" used TDLeaf(lambda) to learn its evaluation function while playing on the Free Internet Chess Server (FICS, The main success we report is that KnightCap improved from a 1650 rating to a 2150 rating in just 308 games and 3 days of play. As a reference, a rating of 1650 corresponds to about level B human play (on a scale from E (1000) to A (1800)), while 2150 is human master level. We discuss some of the reasons for this success, principle among them being the use of on-line, rather than self-play.


page 1

page 2

page 3

page 4


TDLeaf(lambda): Combining Temporal Difference Learning with Game-Tree Search

In this paper we present TDLeaf(lambda), a variation on the TD(lambda) a...

Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm

The game of chess is the most widely-studied domain in the history of ar...

Becoming the World's Highest Rated Chess Player

The Elo rating system measures the approximate skill of each competitor ...

Virtual information system on working area

In order to get strategic positioning for competition in business organi...

HEX and Neurodynamic Programming

Hex is a complex game with a high branching factor. For the first time H...

Manipulating the Distributions of Experience used for Self-Play Learning in Expert Iteration

Expert Iteration (ExIt) is an effective framework for learning game-play...

An Elo-like System for Massive Multiplayer Competitions

Rating systems play an important role in competitive sports and games. T...