| Title: |
Pruning Nodes in the Alpha-Beta Method Using Inductive Logic Programming |
| Authors: |
Nobuhiro Inuzuka; Hayato Fujimoto; Tomofumi Nakano; Hidenori Itoh |
| Contributors: |
The Pennsylvania State University CiteSeerX Archives |
| Source: |
http://www.ai.univie.ac.at/icml-99-ws-games/papers/inuzuka.ps.gz. |
| Collection: |
CiteSeerX |
| Description: |
This paper reports a preliminary research of application of inductive logic programming for pruning nodes in the alpha-beta gametree search method, which finds an appropriate move by looking game-trees in a limited depth ahead. The alpha-beta method reduces the number of nodes by keeping and updating lower and upper bounds of static evaluation. Pruning effect depends on an accidental order of nodes visited, because we can expect large pruning after we have large update. This paper proposes a method to learn rules to sort nodes to yield effective pruning, by using inductive logic programming framework. The method induces a logic program of a binary relation among nodes, and sorts nodes based on the relation. We inspected the method with the game othello. 1 INTRODUCTION Inductive logic programming (ILP) is an inductive learning method in the framework of logic programming, which involves the first-order logic formalism. ILP tries to find a logic program, called a hypothesis, that explai. |
| Document Type: |
text |
| File Description: |
application/postscript |
| Language: |
English |
| Relation: |
http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.50.2311 |
| Availability: |
http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.50.2311; http://www.ai.univie.ac.at/icml-99-ws-games/papers/inuzuka.ps.gz |
| Rights: |
Metadata may be used without restrictions as long as the oai identifier remains attached to it. |
| Accession Number: |
edsbas.BAE48B52 |
| Database: |
BASE |