Learning partially observable action models: efficient algorithms

Authors:
Dafna Shahaf;Allen Chang;Eyal Amir
Affiliations:
Computer Science Department, University of Illinois, Urbana-Champaign, Urbana, IL;Computer Science Department, University of Illinois, Urbana-Champaign, Urbana, IL;Computer Science Department, University of Illinois, Urbana-Champaign, Urbana, IL
Venue:
AAAI'06 Proceedings of the 21st national conference on Artificial intelligence - Volume 1
Year:
2006

Citing 7
Cited 3

Factorial Hidden Markov Models

Machine Learning - Special issue on learning with probabilistic representations
Chaff: engineering an efficient SAT solver

Proceedings of the 38th annual Design Automation Conference
Learning partially observable action schemas

AAAI'06 Proceedings of the 21st national conference on Artificial intelligence - Volume 1
Conformant planning via symbolic model checking

Journal of Artificial Intelligence Research
Planning in nondeterministic domains under partial observability via symbolic model checking

IJCAI'01 Proceedings of the 17th international joint conference on Artificial intelligence - Volume 1
Learning partially observable deterministic action models

IJCAI'05 Proceedings of the 19th international joint conference on Artificial intelligence
Discovering the hidden structure of complex dynamic systems

UAI'99 Proceedings of the Fifteenth conference on Uncertainty in artificial intelligence

Learning action models from plan examples using weighted MAX-SAT

Artificial Intelligence
Learning partially observable deterministic action models

Journal of Artificial Intelligence Research
Learning complex action models with quantifiers and logical implications

Artificial Intelligence

Quantified Score

Hi-index	0.00

Visualization

Abstract

We present tractable, exact algorithms for learning actions' effects and preconditions in partially observable domains. Our algorithms maintain a propositional logical representation of the set of possible action models after each observation and action execution. The algorithms perform exact learning of preconditions and effects in any deterministic action domain. This includes STRIPS actions and actions with conditional effects. In contrast, previous algorithms rely on approximations to achieve tractability, and do not supply approximation guarantees. Our algorithms take time and space that are polynomial in the number of domain features, and can maintain a representation that stays compact indefinitely. Our experimental results show that we can learn efficiently and practically in domains that contain over 1000's of features (more than 21000 states).