Abstract

Intelligent human agents exist in a cooperative social environment that facilitates learning. They learn not only by trialand -error, but also through cooperation by sharing instantaneous information, episodic experience, and learned knowledge. The key investigations of this paper are, "Given the same number of reinforcement learning agents, will cooperative agents outperform independent agents who do not communicate during learning?" and "What is the price for such cooperation?" Using independent agents as a benchmark, cooperative agents are studied in following ways: (1) sharing sensation, (2) sharing episodes, and (3) sharing learned policies. This paper shows that (a) additional sensation from another agent is beneficial if it can be used efficiently, (b) sharing learned policies or episodes among agents speeds up learning at the cost of communication, and (c) for joint tasks, agents engaging in partnership can significantly outperform independent agents although they may learn slo...

Citations

1965	Dynamic Programming - Bellman - 1957
316	Automatic programming of behavior-based robots using reinforcement learning - Mahadevan, Connell - 1992
147	Coordination of Distributed Problem Solvers - Durfee - 1988
79	Toward parallel and distributed learning by metalearning - Chan, Stolfo - 1993
61	Programming robots using reinforcement learning and teaching - Lin - 1991
23	Extending Learning to Multiple Agents : Issues and Model for Multi-Agent, Machine Learning (MA-ML - SIAN
15	A framework for multi-paradigmatic learning - Silver, Frawely, et al. - 1990
12	Cost-sensitive reinforcement learning for adaptive classification and control - Tan - 1991
7	Distributing the planning process in a dynamic environment - Pope, Conry, et al. - 1992
2	A distributed problem-solving approach to rule induction: learning in distributed arti cial intelligence systems - Shaw, Sikora - 1990
1	Integrating agentoriented programming and planning for cooperative problem solving - Tan - 1992