Subgoal Identification for Reinforcement Learning and Planning in Multiagent Problem Solving
- Cite this paper as:
- Chiu CC., Soo VW. (2007) Subgoal Identification for Reinforcement Learning and Planning in Multiagent Problem Solving. In: Petta P., Müller J.P., Klusch M., Georgeff M. (eds) Multiagent System Technologies. MATES 2007. Lecture Notes in Computer Science, vol 4687. Springer, Berlin, Heidelberg
We provide a new probability flow analysis algorithm to automatically identify subgoals in a problem space. Our flow analysis, inspired by preflow-push algorithms, measures the topological structure of the problem space to identify states that connect different subset of state space as the subgoals within linear-time complexity. Then we apply a hybrid approach known as subgoal-based SMDP (semi-Markov Decision Process) that is composed of reinforcement learning and planning based on the identified subgoals to solve the problem in a multiagent environment. The effectiveness of this new method used in a multiagent system is demonstrated and evaluated using a capture-the-flag scenario. We showed also that the cooperative coordination emerged between two agents in the scenario through distributed policy learning.
Unable to display preview. Download preview PDF.