Learning planning rules in noisy stochastic worlds
作者： Leslie Pack Kaelbling , Luke S. Zettlemoyer , Hanna M. Pasula
DOI:
关键词:
摘要: We present an algorithm for learning a model of the effects actions in noisy stochastic worlds. consider 3D simulated blocks world with realistic physics. To this world, we develop planning representation explicit mechanisms expressing object reference and noise. then that can create rules while also derived predicates, evaluate simulator, demonstrating learn effectively dynamics.
参考文章(12)
Herbert A. Simon, Wei-Min Shen, Rule creation and rule learning through environmental exploration international joint conference on artificial intelligence. pp. 675- 680 ,(1989)
Xuemei Wang, Learning by Observation and Practice: An Incremental Approach for Planning Operator Acquisition Machine Learning Proceedings 1995. pp. 549- 557 ,(1995) , 10.1016/B978-1-55860-377-6.50074-8
Avrim L. Blum, John C. Langford, Probabilistic Planning in the Graphplan Framework Lecture Notes in Computer Science. pp. 319- 332 ,(1999) , 10.1007/10720246_25
Michael Kearns, Yishay Mansour, Andrew Y. Ng, A Sparse Sampling Algorithm for Near-Optimal Planning in Large Markov Decision Processes Machine Learning. ,vol. 49, pp. 193- 208 ,(2002) , 10.1023/A:1017932429737
Yolanda Gil, Learning by Experimentation: Incremental Refinement of Incomplete Planning Domains Machine Learning Proceedings 1994. pp. 87- 95 ,(1994) , 10.1016/B978-1-55860-335-6.50019-2
Scott Sherwood Benson, Learning action models for reactive autonomous agents ,(1996)
Leslie P. Kaelbling, Natalia H. Gardiol, Envelope-based Planning in Relational MDPs neural information processing systems. ,vol. 16, pp. 783- 790 ,(2003)
Leslie Pack Kaelbling, Luke S. Zettlemoyer, Hanna M. Pasula, Learning probabilistic relational planning rules principles of knowledge representation and reasoning. pp. 683- 691 ,(2004)
Nada Lavrac, Saso Dzeroski, Inductive Logic Programming: Techniques and Applications Routledge. ,(1993)
Eyal Amir, Learning partially observable deterministic action models international joint conference on artificial intelligence. pp. 1433- 1439 ,(2005)