Abstract
Coordination is an essential technique in cooperative, distributed multiagent systems. However, sophisticated coordination strategies are not always cost-effective in all problem-solving situations. This paper presents a learning method to identify what information will improve coordination in specific problem-solving situations. Learning is accomplished by recording and analyzing traces of inferences after problem solving. The analysis identifies situations where inappropriate coordination strategies caused redundant activities, or the lack of timely execution of important activities, thus degrading system performance. To remedy this problem, situation-specific control rules are created which acquire additional nonlocal information about activities in the agent networks and then select another plan or another scheduling strategy. Examples from a real distributed problem-solving application involving diagnosis of a local area network are described.
Article PDF
Similar content being viewed by others
Avoid common mistakes on your manuscript.
References
Bazzan, A.L.C., Lesser, V.R., & Xuan, P. (1998). Adapting an organizational design through domain-independent diagnosis. (Computer Science Technical Report 98–14). Amherst: University of Massachusetts.
Case, J., Fedor, M., Schoffstall, M., & Davin, J. (1990). A simple network management protocol (SNMP). RFC1157.
Chandy, K.M., & Lamport, L. (1985). Distributed snapshots: Determining global states of distributed systems. ACM Trans. of Computer Systems, 3(1), 63–75.
Corkill, D.D. & Lesser, V.R. (1983). The use of meta-level control forcoordination in a distributed problem solving network (long paper). Proceedings of the Eighth International Joint Conference on Artificial Intelligence (pp. 748–756). Karlsruhe, FRG. (Also published in (1986) Benjamin W. Wah & G.-J. Li (Eds.), Computer architectures for artificial intelligence applications. IEEE Computer Society Press.)
Decker, K., & Lesser, V.R.(1993). Quantitative modeling of complex environments. International Journal of Intelligent Systems in Accounting, Finance and Management, special issue on Mathematical and Computational Models of Organizations: Models and Characteristics of Agent Behavior, 2, 215–234.
Decker, K.S., & Lesser, V.R. (1995). Designing a family of coordination algorithms.Proceedings of the First International Conference on Multi-Agent Systems. San Francisco: AAAI Press.
DeJong, G.(1981). Generalizations based on explanations. Proceedings of Seventh International Joint Conference on Artificial Intelligence (pp. 67–69).
Durfee, E.H., & Lesser, V.R. (1991). Partial global planning: A coordination framework fordistributed hypothesis formation. IEEE Transactions on Systems, Man, and Cybernetics, 21(5), 1167–1183.
Durfee, E.H., Lesser, V.R., & Corkill, D.D. (1987). Coherent cooperation among communicating problem solvers. IEEETransactions on Computers, 36(11), 1275–1291. (Also published in (1988) A. Bond & L. Gasser (Eds.), Readings in distributed artificial intelligence. CA: Morgan Kaufmann Publishers.)
Grefenstette, J. (1992). The evolutionof strategies for multi-agent environments. Adaptive Behavior, 1(1), 65–89.
Hammond, K.J. (1989). Casebasedplanning: Viewing planning as a memory task. Academic Press.
Hudlickà, E., & Lesser, V.R. (1984). Meta-levelcontrol through fault detection and diagnosis. Proceedings of the 1984 National Conference on AI (pp. 153–161).
Hudlickà, E., & Lesser, V.R. (1987). Modeling and diagnosing problem-solving system behavior. IEEETransactions on Systems, Man, and Cybernetics, 17(3), 407–419. (Also published in (1988) A. Bond & L. Gasser (Eds.), Readings in distributed artificial intelligence. CA:Morgan Kaufmann Publishers.)
Kinney, M., & Tsatsoulis, C. (1993).Learning communication strategies in distributed agent environments. (Working Paper CECASE-WP-93-4). Lawrence: Center for Excellence in Computer-Aided Systems Engineering, The University of Kansas.
Lesser, V.R. (1991).A retrospective view of FA/C distributed problem solving. IEEE Transactions on Systems, Man, and Cybernetics, 21(6), 1347–1362.
Lesser, V.R. (1998). Reflections of the nature of multi-agent coordination and its implications for an agentarchitecture. Autonomous Agents and Multi-Agent Systems, 1, 89–111.
Lesser, V.R., Decker, K., Carver, N., Garvey, A., Neiman, D., Nagendra Prasad, M., & Wagner, T. (1998). Evolution of the GPGP domain-independent coordination framework. (Computer Science Technical Report 98-05). Amherst: University of Massachusetts.
Lesser, V., Nawab, H., & Klassner, F. (1995). IPUS:Anarchitecture for the integrated processing and understanding of signals.Artificial Intelligence, 77, 129–171.
Nagendra Prasad, M.V., & Lesser, V. (1996). Off-line learning ofcoordination in functionally structured agents for distributed data processing. Proceedings of the ICMAS-96 Workshop on LIOME. (An extended version of the paper appears as: (1997). Learning situation-specific coordination in cooperative multi-agent systems. (Computer Science Technical Report 97-12). Amherst: University of Massachusetts.
NagendraPrasad, M.V., Lesser, V.R., & Lander, S. (1998). Learning organizational roles for negotiated search in a multi-agent system. International Journal of Human-Computer Studies (IJHCS), special issue on Evolution and Learning in Multi-Agent Systems, 48, 51–67.
Mitchell, T.M., Keller, R.M., & Kedar-Cabelli, S.T. (1986). Explanation-based generalizations: Aunifying view. Machine Learning, 1, 47–80.
Sandholm, T., & Crites, R. (1995). Multi-agent reinforcementlearning in the iterated prisoner's dilemma. Biosystems, 37, 147–166, special issue on the Prisoner's Dilemma.
Sen, S., Sekaran, M., & Hale, J. (1994). Learning to coordinate without sharing information. Proceedings of the 1994National Conference on AI (pp. 426–431).
Shoham, Y., & Tennenholtz, M. (1992). Emergent conventions in multi-agentsystems: Initial experimental results and observations. Proceedings of KR-92.
Sugawara, T. (1990). A cooperative LANdiagnostic and observation expert system. Proceedings of IEEE Phoenix Conference on Comp. and Comm. (pp. 667–674).
Sugawara, T., & Lesser, V.R. (1993). On-line learning of coordination plans. (Computer Science Technical Report93-27). Amherst: University of Massachusetts. (A shorter version of this paper was also published in (1993) Proc. of the 12th Intl. Workshop on Distributed AI.)
Sugawara, T., & Murakami, K. (1992). A multiagent diagnostic system forinternetwork problems. Proceedings of INET'92.
Sycara, K. (1989). Multiagent compromise via negotiation. InL. Gasser & M.N. Huhns (Eds.), Distributed artificial intelligence II. CA: Morgan Kaufmann Publishers.
Tan, M.(1993). Multi-agent reinforcement learning: Independent vs. cooperative agents. Proceedings of the Tenth International Conference on Machine Learning (pp. 330–337).
Weiss, G. (1994). Some studies in distributed machine learning andorganizational design. (Technical Report FKI-189-94). TU, Mnchen: Institut für Informatik.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Sugawara, T., Lesser, V. Learning to Improve Coordinated Actions in Cooperative Distributed Problem-Solving Environments. Machine Learning 33, 129–153 (1998). https://doi.org/10.1023/A:1007510522680
Issue Date:
DOI: https://doi.org/10.1023/A:1007510522680