Using patterns and plans in chess

doi:10.1016/0004-3702(80)90039-9

Artificial Intelligence

Volume 14, Issue 2, September 1980, Pages 165-203

https://doi.org/10.1016/0004-3702(80)90039-9 Get rights and content

Abstract

The purpose of this research is to investigate the extent to which knowledge can replace and support search in selecting a chess move and to delineate the issues involved. This has been carried out by constructing a program, PARADISE (PAttern Recognition Applied to DIrecting SEarch), which finds the best move in tactically sharp middle game positions from the games of chess masters. It encodes a large body of knowledge in the form of production rules. The actions of the rules post concepts in the data base while the conditions match patterns in the chess position and data base. The program uses the knowledge base to discover plans during static analysis and to guide a small tree search which confirms that a particular plan is best. The search is “small” in the sense that the size of the search tree is of the same order of magnitude as a human master's search tree (tens and hundreds of nodes, not thousands to hundreds of thousands as in many computer chess programs).

Once a plan is formulated, it guides the tree search for several ply and expensive static analyses (needed to analyze a new position) are done infrequently. PARADISE avoids placing a depth limit on the search (or any other artificial effort limit). By using a global view of the search tree, information gathered during the search, and the analysis provided by the knowledge base, the program produces enough terminations to force convergence of the search. PARADISE has found combinations as deep as 19 ply.

References (14)

J. Gillogly
The technology chess program
Artificial Intelligence
(1972)
J. Pitrat
A chess combination program which uses plans
Artificial Intelligence
(1977)
E.D. Sacerdoti
Planning in a hierarchy of abstraction spaces
Artificial Intelligence
(1974)
H. Berliner
Chess as problem solving: The development of a tactics analyzer
H. Charness
Human chess skill
A.D. De Groot
R. Davis
Applications of meta level knowledge to the construction, maintenance and use of large knowledge bases

There are more references available in the full text version of this article.

Cited by (46)

The emergence of choice: Decision-making and strategic thinking through analogies
2014, Information Sciences
Citation Excerpt :
Moreover, they are hardly verbalizable: because relations such as “attack” can be stored over trajectories with many movements, it is a matter of context whether or not a piece is placing pressure on another piece or in a particular area). In this section we describe an actual execution of Capyblanca in problem 6, a variation of Wilkins’ [42] position. This is a much harder position, in fact, than that used by Wilkins, with increased search depth and breadth.
Consider the chess game: When faced with a complex scenario, how does understanding arise in one’s mind? How does one integrate disparate cues into a global, meaningful whole? How do players avoid the combinatorial explosion? How are abstract ideas represented? The purpose of this paper is to propose a new computational model of human chess cognition. We suggest that analogies and abstract roles are crucial to understanding a chess scenario. We present a proof-of-concept model, in the form of a computational architecture, which accounts for many crucial aspects of human play, such as (i) concentration of attention to relevant aspects, (ii) how humans may avoid the combinatorial explosion, (iii) perception of similarity at a strategic level, (iv) a state of meaningful anticipation over how a global scenario may evolve, and (v) the architecture’s choice as an emergent phenomenon from the actions of subcognitive processes.
Applying adversarial planning techniques to Go
2001, Theoretical Computer Science
Approaches to computer game playing based on $α$ – $β$ search of the tree of possible move sequences combined with a position evaluation function have been successful for many games, notably Chess. Such approaches are less successful for games with large search spaces and complex positions, such as Go, and we are led to seek alternatives. One such alternative is to model the goals of the players, and their strategies for achieving these goals. This approach means searching the space of possible goal expansions, typically much smaller than the space of move sequences. Previous attempts to apply these techniques to Go have been unable to provide results for anything other than a high strategic level or very open game positions. In this paper we describe how adversarial hierarchical task network planning can provide a framework for goal-directed game playing in Go which is also applicable both strategic and tactical problems.
Computer chess move-ordering schemes using move influence
2000, Artificial Intelligence
The chessmaps heuristic is a pattern-oriented approach to ordering moves for the game of chess. It uses a neural network to learn a relation between the control of the squares and the influence of a move. Depending on what squares a player controls, the chessmaps heuristic tries to determine where the important areas of the chessboard are. Moves that influence these important areas are then ordered first. The heuristic has been incorporated into a move-ordering algorithm that also takes account of immediate tactical threats. Human players also rely strongly on patterns when selecting moves, but would also consider immediate tactical threats, so this move-ordering algorithm is an attempt to mimic something of the human thought process when selecting a move. This paper presents a new definition for the influence of a move, which improves the performance of the heuristic. It also presents a new experience-based approach to determining what areas of the chessboard are important, which may actually be preferred to the chessmaps heuristic. The results from game-tree searches suggest that the move-ordering algorithm could compete with the current best alternative of using the history heuristic with capture moves in a brute-force search.
Is artificial intelligence a degenerating program?: a review of Hubert Dreyfus' what computers still can't do
1996, Artificial Intelligence
Subgoal search for complex reasoning tasks
2021, arXiv
Subgoal Search For Complex Reasoning Tasks
2021, Advances in Neural Information Processing Systems

View all citing articles on Scopus

View full text

Using patterns and plans in chess

Abstract

Artificial Intelligence

Artificial Intelligence

Artificial Intelligence

Chess as problem solving: The development of a tactics analyzer

Human chess skill

Applications of meta level knowledge to the construction, maintenance and use of large knowledge bases