Towards Social Complexity Reduction in Multiagent Learning: the ADHOC Approach

This paper presents a novel method for classifying adversaries that is designed to achieve social complexity reduction in large-scale, open multiagent systems. In contrast to previous work on opponent modelling, we seek to generalise from individuals and to identify suitable opponent classes. To validate the adequacy of our approach, we present initial experiments in a multiagent Iterated Prisoner’s Dilemma seenario and we discuss directions for future work on the subject.

[1]  Michael Rovatsos,et al.  Hierarchical common-sense interaction learning , 2000, Proceedings Fourth International Conference on MultiAgent Systems.

[2]  Thomas G. Dietterich.  What is machine learning? , 2020, Archives of Disease in Childhood.

[3]  Peter Dayan,et al.  Q-learning , 1992, Machine Learning.

[4]  Stuart J. Russell,et al.  Do the right thing - studies in limited rationality , 1991 .

[5]  David Carmel,et al.  Learning and using opponent models in adversary search , 1996 .

[6]  Edmund H. Durfee,et al.  Agents Learning about Agents: A Framework and Analysis , 1997 .

[7]  Peter Stone,et al.  Layered learning in multiagent systems - a winning approach to robotic soccer , 2000, Intelligent robotics and autonomous agents.

[8]  W. Hamilton,et al.  The evolution of cooperation. , 1984, Science.

[9]  Ronitt Rubinfeld,et al.  Efficient algorithms for learning to play repeated games against computationally bounded adversaries , 1995, Proceedings of IEEE 36th Annual Foundations of Computer Science.

[10]  Tucker Balch,et al.  Learning Roles: Behavioral Diversity in Robot Teams , 1997 .

[11]  Svetha Venkatesh,et al.  Learning other Agents' Preferences in Multiagent Negotiation , 1996, AAAI/IAAI, Vol. 1.

[12]  David Carmel,et al.  Learning Models of Intelligent Agents , 1996, AAAI/IAAI, Vol. 1.