Advertisement

Extraction of Implicit Resource Relationships in Multi-agent Systems

  • Toshiharu Sugawara
  • Satoshi Kurihara
  • Osamu Akashi
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2891)

Abstract

This paper discusses the storage and analysis of past hierarchical-planning results in order to identify implicit costs and resource relationships between activities in multi-agent contexts. We have previously proposed a plan-reuse framework in which plans are stored as templates after use and then reused to speed up planning activity in multi-agent systems. In this paper, we propose the mechanizm for learning, from templates that consist of used plans and data recorded during planning and execution, implicit relationships concerning resource usage by multiple agents. Here, implicit indicates that the relationships exist in the environments where agents are deployed but are not described in the domain models the agents have. The plan-reuse framework also provides guidance on which data the planner and executor should record and on when the learned rules should be applied. Finally, some examples show how this learning enables the creation of more appropriate solutions by agents.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Allen, J.F.: Maintaining knowledge about temporal intervals. Communications of the ACM 26(11), 832–843 (1983)zbMATHCrossRefGoogle Scholar
  2. 2.
    Alterman, R., Garland, A.: Convention in Joint Activity. Cognitive Science 25(4), 611–657 (2001)CrossRefGoogle Scholar
  3. 3.
    Erol, K., Hendler, J., Nau, D.S.: HTN planning: Complexity and expressivity. In: Proc. of the AAAI 1994, vol. 2, pp. 1123–1128 (1994)Google Scholar
  4. 4.
    Georgeff, M.P., Lansky, A.L.: Procedural knowledge. In: Proc. of the IEEE Special Issue on Knowledge Representation, vol. 74, pp. 1383–1398 (1986)Google Scholar
  5. 5.
    Horling, B., Benyo, B., Lesser, V.: Using self-diagnosis to adapt organizational structures. In: Proc. of the Int. Conf. on Autonomous Agents, pp. 529–536. ACM Press, New York (2001)CrossRefGoogle Scholar
  6. 6.
    Kambhampati, S.: Supporting flexible plan reuse. In: Minton, S. (ed.) Machine Learning Methods for Planning, pp. 397–434. Morgan Kaufmann, San Francisco (1993)Google Scholar
  7. 7.
    Prasad, M.N., Lesser, V.R.: The use of meta-level information in learning situation specific coordination. In: Proc. of the IJCAI 1997, pp. 640–646 (1997)Google Scholar
  8. 8.
    Sacerdoti, E.: Planning in a hierarchy of abstraction spaces. Artificial Intelligence 5(2), 115–135 (1974)zbMATHCrossRefGoogle Scholar
  9. 9.
    Sugawara, T.: Reusing past plans in distributed planning. In: Proc. of the Int. Conf. on Multi- Agent Systems (ICMAS 1995), pp. 360–367 (1995)Google Scholar
  10. 10.
    Sugawara, T., Lesser, V.: Learning to improve coordinated actions in cooperative distributed problem-solving environments. Machine Learning 33(2/3), 129–153 (1998)zbMATHCrossRefGoogle Scholar
  11. 11.
    Wooldridge, M.: An Introduction to MultiAgent Systems. John Wiley & Sons, Chichester (2002)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2003

Authors and Affiliations

  • Toshiharu Sugawara
    • 1
  • Satoshi Kurihara
    • 1
  • Osamu Akashi
    • 1
  1. 1.NTT Network Innovations LaboratoriesTokyoJapan

Personalised recommendations