Home | Publications | CS Home

Learning to Play Chess Selectively by Acquiring Move Patterns

---

Lev Finkelstein and Shaul Markovitch. Learning to Play Chess Selectively by Acquiring Move Patterns. ICCA Journal, 21:100-119 1998.

---

Abstract

Several researchers have noted that human chess players do not perceive a position as a static entity, but as a collection of potential actions. Indeed, it looks as if human chess players are able to follow promising moves without considering all the alternatives. This work studies the possibility of incorporating such capabilities into chess programs. We present a methodology for representing move patterns. A move pattern is a structure consisting of a board pattern and a move that can be applied in that pattern. Move patterns are used for selecting promising branches of the search tree, allowing a narrower, and therefore deeper, search. Move patterns are learned during training games and are stored in an hierarchical structure to enable fast retrieval. The paper describes a language for representing move patterns, and algorithms for learning, storing, retrieving and using them.

---

Keywords: Learning in Games, Games, Relational Reinforcement Learning

---

Secondary Keywords:

---

Online version:

---

Bibtex entry:

 @article{Finkelstein:1998:LPC,
  Author = {Lev Finkelstein and Shaul Markovitch},
  Title = {Learning to Play Chess Selectively by Acquiring Move Patterns},
  Year = {1998},
  Journal = {ICCA Journal},
  Volume = {21},
  Number = {2},
  Pages = {100--119},
  Url = {http://www.cs.technion.ac.il/~shaulm/papers/pdf/Finkelstein-Markovitch-icca1998.pdf},
  Keywords = {Learning in Games, Games, Relational Reinforcement Learning},
  Secondary-keywords = {Pattern Learning, Explanation-Based Learning, Adversary Search},
  Abstract = {
    Several researchers have noted that human chess players do not
    perceive a position as a static entity, but as a collection of
    potential actions. Indeed, it looks as if human chess players are
    able to follow promising moves without considering all the
    alternatives. This work studies the possibility of incorporating
    such capabilities into chess programs. We present a methodology
    for representing move patterns. A move pattern is a structure
    consisting of a board pattern and a move that can be applied in
    that pattern. Move patterns are used for selecting promising
    branches of the search tree, allowing a narrower, and therefore
    deeper, search. Move patterns are learned during training games
    and are stored in an hierarchical structure to enable fast
    retrieval. The paper describes a language for representing move
    patterns, and algorithms for learning, storing, retrieving and
    using them.
  }

  }