Search
  • All Journals
Advanced Search
Search articles by Title, Author, Keywords, DOI
3.9CiteScore
1.3Impact Factor
Advanced Search
MAP  >  The Knowledge Engineering Review
2007 Volume 22
Article Contents
Next Previous
RESEARCH ARTICLE   Open Access    

ARMS: an automatic knowledge engineering tool for learning action models for AI planning

More Information
  • Abstract: We present an action model learning system known as ARMS (Action-Relation Modelling System) for automatically discovering action models from a set of successfully observed plans. Current artificial intelligence (AI) planners show impressive performance in many real world and artificial domains, but they all require the definition of an action model. ARMS is aimed at automatically learning action models from observed example plans, where each example plan is a sequence of action traces. These action models can then be used by the human editors to refine. The expectation is that this system will lessen the burden of the human editors in designing action models from scratch. In this paper, we describe the ARMS in detail. To learn action models, ARMS gathers knowledge on the statistical distribution of frequent sets of actions in the example plans. It then builds a weighted propositional satisfiability (weighted SAT) problem and solves it using a weighted MAXSAT solver. Furthermore, we show empirical evidence that ARMS can indeed learn a good approximation of the finally action models effectively.
  • 加载中
  • Agrawal Rakesh and Srikant Ramakrishnan, 1994 Fast algorithms for mining association rules. In Proceedings of the 20th International Conference on Very Large Data Bases (VLDB). Morgan Kaufmann, pp. 487–499.

    Google Scholar

    Andrew Garland and Neal Lesh, 2002 Plan evaluation with incomplete action descriptions. In Proceedings of the Seventeenth National Conference on Artificial Intelligence 2002. AAAI, pp. 461–467.

    Google Scholar

    Bacchus Fahiem and Kabanza Froduald, 2000Using temporal logics to express search control knowledge for planning. Artificial Intelligence116123–191.

    Google Scholar

    Benson Scott, 1995 Inductive learning of reactive action models. In International Conference on Machine Learning, pp. 47–54.

    Google Scholar

    Blum Avrim and Furst Merrick, 1997Fast planning through planning graph analysis. Artificial Intelligence90281–300.

    Google Scholar

    Blythe Jim, Kim Jihie, Ramachandran Surya and Gil Yolanda, 2001 An integrated environment for knowledge acquisition. In Intelligent User Interfaces. ACM, pp. 13–20.

    Google Scholar

    Borchers Brian and Furman Judith, 1999A two-phase exact algorithm for max-sat and weighted max-sat problems. Journal of Combinatorial Optimization2(4) 299–306.

    Google Scholar

    Bresina John, Jonsson Ari, Morris Paul and Rajan Kanna, 2005 Activity planning for the mars exploration rovers. In Proceedings of the Fifteenth International Conference on Automated Planning and Scheduling (ICAPS). AAAI, pp. 40–49.

    Google Scholar

    Cheeseman Peter, Kanefsky Bob and Taylor William M., 1991 Where the really hard problems are. In Proceedings of the Seventh International Joint Conference on Artificial Intelligence (IJCAI). Morgan Kaufmann, pp. 331–337.

    Google Scholar

    Edelkamp Stefan, and Mehler Tilman, 2005 Knowledge acquisition and knowledge engineering in the modplan workbench. In Proceedings of the Fifteenth International Conference on Automated Planning and Scheduling (ICAPS). AAAI, pp. 26–33.

    Google Scholar

    Fikes Richard E. and Nilsson Nils J., 1971Strips: A new approach to the application of theorem proving to problem solving. Artificial Intelligence2189–208.

    Google Scholar

    Fink Eugene and Yang Qiang, 1997Automatically selecting and using primary effects in planning: Theory and experiments. Artificial Intelligence89(1–2) 285–315.

    Google Scholar

    Fox Maria and Long Derek, 2003PDDL2.1: An extension to PDDL for expressing temporal planning domains. Journal of Artificial Intelligence Research2061–124.

    Google Scholar

    Garland Andrew and Lesh Neal, 2002 Plan evaluation with incomplete action descriptions. In Proceedings of the Eighteenth National Conference on AI (AAAI 2002). AAAI, pp. 461–467.

    Google Scholar

    Gil Yolanda, 1994 Learning by experimentation: Incremental refinement of incomplete planning domains. In Eleventh Intl Conf on Machine Learning. Morgan Kaufmann, pp. 87–95.

    Google Scholar

    Kautz Henry and Selman Bart, 1996 Pushing the envelope: Planning, propositional logic, and stochastic search. In Proceedings of the Thirteenth National Conference on Artificial Intelligence(AAAI). AAAI, pp. 1194–1201.

    Google Scholar

    Kautz Henry A. and Allen James F., 1986 Generalized plan recognition. In Proceedings of the Fifth National Conference on Artificial Intelligence (AAAI). AAAI, pp. 32–37.

    Google Scholar

    McCluskey Thomas Leo, Liu D. and Simpson Ron M., 2003 Gipo ii: Htn planning in a tool-supported knowledge engineering environment. In Proceedings of the International Conference on Automated Planning and Scheduling(ICAPS). AAAI, pp. 92–101.

    Google Scholar

    McCluskey Thomas Leo,Richardson N. and Simpson Ron M., 2002 An Interactive Method for Inducing Operator Descriptions. In Proceedings of the 6th International Conference on AI Planning and Scheduling (AIPS-2002). AAAI.

    Google Scholar

    Moskewicz Matthew W., Madigan Conor F., Zhao Ying, Zhang Lintao and Malik Sharad, 2001 Chaff: Engineering an efficient sat solver. In Proceedings of the 38th Design Automation Conference (DAC). ACM.

    Google Scholar

    Myers Karen L., 2005 Metatheoretic Plan Summarization and Comparison. In Proceedings of the ICAPS-05 Workshop on Mixed-initiative Planning and Scheduling. AAAI.

    Google Scholar

    Nau Dana S., Au Tsz-Chiu, Ilghami Okhtay, Kuter Ugur, William Murdock J., Wu Dan and Yaman Fusun, 2003Shop2: An htn planning system. Journal of Artificial Intelligence Research20379–404.

    Google Scholar

    Nau Dana S., Au Tsz-Chiu, Ilghami Okhtay, Kuter Ugur, William Murdock J., Wu Dan and Yaman Fusun, 2005Applications of shop and shop2. IEEE Intelligent Systems, 20(2) 34–41.

    Google Scholar

    Oates Tim and Cohen Paul R., 1996 Searching for planning operators with context-dependent and probabilistic effects. In Proceedings of the Thirteenth National Conference on AI (AAAI 96). AAAI, pp. 865–868.

    Google Scholar

    Pednault Edwin P. D., 1986 Formulating multiagent, dynamic-world problems in the classical planning framework. In Reasoning about Actions and Plans: Proceedings of the 1986 Workshop. Morgan Kaufmann, pp. 47–82.

    Google Scholar

    Pistore Marco, Traverso Paolo and Bertoli Piergiorgio, 2005 Automated composition of web services by planning in asynchronous domains. In Proceedings of the Fifteenth International Conference on Automated Planning and Scheduling(ICAPS). AAAI, pp. 2–11.

    Google Scholar

    Sablon Gunther and Boulanger Dmitri, 1994 Using the event calculus to intetgrate planning and learning in an intelligent autonomous agent. In Current Trends in AI Planning. IOS Press, pp. 254–265.

    Google Scholar

    Shen Weimin, 1994Autonomous Learning from the Environment. Computer Science Press, W.H. Freeman and Company.

    Google Scholar

    Wang Xuemei, 1995 Learning by observation and practice: An incremental approach for planning operator acquisition. In Proceedings of the Twelfth International Conference on Machine Learning(ICML). Morgan Kaufmann, pp. 549–557.

    Google Scholar

    Winner Elly and Veloso Manuela, 2002 Analyzing plans with conditional effects. In Proceedings of the Sixth International Conference on AI Planning and Scheduling(AIPS).AAAI.

    Google Scholar

    Yang Qiang, Wu Kangheng and Jiang Yunfei, 2005 Learning action models from plan examples with incomplete knowledge. In Proceedings of the Fifteenth International Conference on Automated Planning and Scheduling(ICAPS). AAAI, pp. 241–250.

    Google Scholar

    Younes Haakan L. S. and Littman Michael L., 2004 PPDDL1.0: An Extension to PDDL for Expressing Planning Domains with Probabilistic Effects. In CMU-CS-04-167, Carnegie Mellon University.

    Google Scholar

    Wu D., Sirin E., Hendler J., Nau D. and Parsia B., 2003 Automatic web services composition using SHOP2. In Twelfth International World Wide Web Conference (WWW2003). ACM.

    Google Scholar

    Zhang Hantao, 1997 SATO: an efficient propositional prover. In Proceedings of the International Conference on Automated Deduction (CADE). Springer, pp. 272–275.

    Google Scholar

  • Cite this article

    KANGHENG WU, QIANG YANG, YUNFEI JIANG. 2007. ARMS: an automatic knowledge engineering tool for learning action models for AI planning. The Knowledge Engineering Review. 22:87 doi: 10.1017/S0269888907001087
    KANGHENG WU, QIANG YANG, YUNFEI JIANG. 2007. ARMS: an automatic knowledge engineering tool for learning action models for AI planning. The Knowledge Engineering Review. 22:87 doi: 10.1017/S0269888907001087
    shu
Download PDF

Article Metrics

Article views(28) PDF downloads(317)

Other Articles By Authors

RESEARCH ARTICLE   Open Access    

ARMS: an automatic knowledge engineering tool for learning action models for AI planning

  • Published online: 01 June 2007
The Knowledge Engineering Review  22,  Article number: 10.1017/S0269888907001087  (2007)  |  Cite this article

Abstract: Abstract: We present an action model learning system known as ARMS (Action-Relation Modelling System) for automatically discovering action models from a set of successfully observed plans. Current artificial intelligence (AI) planners show impressive performance in many real world and artificial domains, but they all require the definition of an action model. ARMS is aimed at automatically learning action models from observed example plans, where each example plan is a sequence of action traces. These action models can then be used by the human editors to refine. The expectation is that this system will lessen the burden of the human editors in designing action models from scratch. In this paper, we describe the ARMS in detail. To learn action models, ARMS gathers knowledge on the statistical distribution of frequent sets of actions in the example plans. It then builds a weighted propositional satisfiability (weighted SAT) problem and solves it using a weighted MAXSAT solver. Furthermore, we show empirical evidence that ARMS can indeed learn a good approximation of the finally action models effectively.

    HTML

    Rights and permissions
    • Copyright © Cambridge University Press 20072007Cambridge University Press
References (34)
  • About this article
    Cite this article
    KANGHENG WU, QIANG YANG, YUNFEI JIANG. 2007. ARMS: an automatic knowledge engineering tool for learning action models for AI planning. The Knowledge Engineering Review. 22:87 doi: 10.1017/S0269888907001087
    KANGHENG WU, QIANG YANG, YUNFEI JIANG. 2007. ARMS: an automatic knowledge engineering tool for learning action models for AI planning. The Knowledge Engineering Review. 22:87 doi: 10.1017/S0269888907001087
    shu

    • Catalog

      /

      DownLoad:  Full-Size Img  PowerPoint
      Return
      Return