The Sage development method and associated tool set support an incremental, iterative, model-driven process to build and maintain high assurance, reactive multi-agent systems. A s...
Bayesian inference is an appealing approach for leveraging prior knowledge in reinforcement learning (RL). In this paper we describe an algorithm for discovering different classes...
Abstract Many Multi-Agent Systems (MAS) methodologies incorporate a modeldriven approach. Model Driven Engineering is based on three main ideas: models are the “first-class citi...
Situated Multi-Agents Systems (MAS), and other Agentbased systems, are often complex. Formal reasoning is needed to ensuring their correctness and structuring their development. E...
Recent multi-agent extensions of Q-Learning require knowledge of other agents’ payoffs and Q-functions, and assume game-theoretic play at all times by all other agents. This pap...