Skip To Content

Athabasca University

Section 3 : Multiagent learning

Commentary

Section Goals

  • To discuss how machine learning performs in a multiagent environment.

Learning Objectives

Learning Objective 1

  • Outline the problems and methods of multiagent learning.
  • Explain the principles of multiagent learning compared to single agent learning presented in this unit.
  • Discuss the benefits and challenges of multiagent learning.

Objective Readings

Required readings:

Vohra, R., & Wellman, M. (eds.). (2007). Special Issue on Multi-Agent Learning. Artificial Intelligence, 171(7), 363-452.

Begin by reading the first paper, and then browse through several other papers. Try to find one or two papers that are of the greatest interest to you, and read them carefully.

  • Vohra, R. V., and Wellman, M. P. (2007). Foundations of multi-agent learning: Introduction to the special issue. Special Issue on Multi-Agent Learning in Artificial Intelligence, 171(7), 363-364.
  • Shoham, Y., Powers, R., and Grenager, T. (2007). If multi-agent learning is the answer, what is the question? Special Issue on Multi-Agent Learning in Artificial Intelligence, 171(7), 365-377.
  • Fudenberg, D., and Levine, D. K. (2007). An economist's perspective on multi-agent learning. Special Issue on Multi-Agent Learning in Artificial Intelligence, 171(7), 378-381.
  • Sandholm, T. (2007). Perspectives on multiagent learning. Special Issue on Multi-Agent Learning in Artificial Intelligence, 171(7), 382-391.
  • Gordon, G. J. (2007). Agendas for multi-agent learning. Special Issue on Multi-Agent Learning in Artificial Intelligence, 171(7), 392-401.
  • Stone, P. (2007). Multiagent learning is not the answer. It is the question. Special Issue on Multi-Agent Learning in Artificial Intelligence, 171(7), 402-405.
  • Tuyls, K., and Parsons, S. (2007). What evolutionary game theory tells us about multiagent learning. Special Issue on Multi-Agent Learning in Artificial Intelligence, 171(7), 406-416.
  • Mannor, S., and Shamma, J. S. (2007). Multi-agent learning for engineers. Special Issue on Multi-Agent Learning in Artificial Intelligence, 171(7), 417-422.
  • Erev, I., and Roth, A. E. (2007). Multi-agent learning and the descriptive value of simple models. Special Issue on Multi-Agent Learning in Artificial Intelligence, 171(7), 423-428.
  • Young, H. P. (2007). The possible and the impossible in multi-agent learning. Special Issue on Multi-Agent Learning in Artificial Intelligence, 171(7), 429-433.
  • Chang, Y.-H. (2007). No regrets about no-regret. Special Issue on Multi-Agent Learning in Artificial Intelligence, 171(7), 434-439.
  • Zinkevich, M., Greenwald, A., and Littman, M. L. (2007). A hierarchy of prescriptive goals for multiagent learning. Special Issue on Multi-Agent Learning in Artificial Intelligence, 171(7), 440-447.
  • Monderer, D., and Tennenholtz, M. (2007). Learning equilibrium as a generalization of learning to optimize. Special Issue on Multi-Agent Learning in Artificial Intelligence, 171(7), 448-452.

Other Papers:

Choose one long technical paper about multiagent learning from the main AI journals, preferably from the Reference section in one of the papers listed above.

Objective Questions

  • What do you think is one of the best test beds for multiagent learning research?
  • Is there a general guideline for expending single agent learning methods to multiagent learning?

Objective Activities

  • Explore multiagent learning applications or prototype systems for further research.

Updated December 14 2017 by FST Course Production Staff