The Complexity of Decentralized Control of Markov Decision Processes
Mathematics of Operations Research
On the relationship between MDPs and the BDI architecture
AAMAS '06 Proceedings of the fifth international joint conference on Autonomous agents and multiagent systems
Programming Multi-Agent Systems in AgentSpeak using Jason (Wiley Series in Agent Technology)
Programming Multi-Agent Systems in AgentSpeak using Jason (Wiley Series in Agent Technology)
Cooperative Interactions: An Exchange Values Model
Coordination, Organizations, Institutions, and Norms in Agent Systems II
Hybrid BDI-POMDP framework for multiagent teaming
Journal of Artificial Intelligence Research
Planning and acting in partially observable stochastic domains
Artificial Intelligence
Coordination and sociability for intelligent virtual agents
COIN'07 Proceedings of the 2007 international conference on Coordination, organizations, institutions, and norms in agent systems III
Hi-index | 0.01 |
This paper introduces an approach for the self-regulation of personality-based social exchange processes in multiagent systems, where a mechanism of social equilibrium supervision is internalized in the agents with the goal of achieving the equilibrium of the exchanges, guaranteeing the continuity of the interactions. The decision process concerning the best exchanges that an agent should propose to its partner is modeled as one Partially Observable Markov Decision Processs (POMDPs) for each agent personality trait. Each POMDP is decomposed into sub-POMDPs, according to the current balance of exchange values. Based on the relationship that may be established between POMDPs and Belief-Desire-Intention (BDI) architectures, the paper introduces an algorithm to map the policy graphs of the sub-POMDPs to BDI plans, allowing for a library of BDI plans. Simulations were carried out considering different degrees of supervision.