Learning Game-Specific Spatially-Oriented Heuristics

Authors:
Susan L. Epstein;Jack Gelfand;Esther Lock
Affiliations:
Department of Computer Science, Hunter College and The Graduate School, The City University of New York, New York, NY 10021;Department of Psychology, Princeton University, Princeton, NJ 08544;Department of Computer Science, The Graduate School of The City University of New York, New York, NY 10036
Venue:
Constraints
Year:
1998

Citing 9
Cited 6

Features and objects in visual processing

Scientific American
Toward an Ideal Trainer

Machine Learning
Methods for combining experts' probability assessments

Neural Computation
Some studies in machine learning using the game of checkers

Computers & thought
The sciences of the artificial (3rd ed.)

The sciences of the artificial (3rd ed.)
Empirical Support for Winnow and Weighted-MajorityAlgorithms: Results on a Calendar Scheduling Domain

Machine Learning
Spatial Reasoning for the Automatic Recognition of Machinable Features in Solid Models

IEEE Transactions on Pattern Analysis and Machine Intelligence
Learning Quickly When Irrelevant Attributes Abound: A New Linear-Threshold Algorithm

Machine Learning
Constructive Function Approximation TITLE2:

Constructive Function Approximation TITLE2:

Game playing (invited talk): the next moves

AAAI '99/IAAI '99 Proceedings of the sixteenth national conference on Artificial intelligence and the eleventh Innovative applications of artificial intelligence conference innovative applications of artificial intelligence
Learning to play strong poker

Machines that learn to play games
Learning on Paper: Diagrams and Discovery in Game Playing

DIAGRAMS '02 Proceedings of the Second International Conference on Diagrammatic Representation and Inference
Collaborative Learning for Constraint Solving

CP '01 Proceedings of the 7th International Conference on Principles and Practice of Constraint Programming
The Adaptive Constraint Engine

CP '02 Proceedings of the 8th International Conference on Principles and Practice of Constraint Programming
Protein Structure from Contact Maps: A Case-Based Reasoning Approach

Information Systems Frontiers

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper describes an architecture that begins withenough general knowledge to play any board game as a novice,and then shifts its decision-making emphasis to learned, game-specific,spatially-oriented heuristics. From its playing experience, itacquires game-specific knowledge about both patterns and spatialconcepts. The latter are proceduralized as learned, spatially-orientedheuristics. These heuristics represent a new level of featureaggregation that effectively focuses the program‘s attention.While training against an external expert, the program integratesthese heuristics robustly. After training it exhibits both anew emphasis on spatially-oriented play and the ability to respondto novel situations in a spatially-oriented manner. This significantlyimproves performance against a variety of opponents. In addition,we address the issue of context on pattern learning. The proceduresdescribed here move toward learning spatially-oriented heuristicsfor autonomous programs in other spatial domains.