Un modèle de mémoire pour l'apprentissage de communication dans un SMA

Learning in a multi-agent context is a difficult task, especially since the communication with other agents require additional information to be stored and used in the learning process. In this paper, we propose a model of agent memory for command and control in a multi-agent context. We first show why the existing memory models are not sufficient to support automated learning of command and control in a multi-agent context. Then, we describe our model for storing messages and answers. Finally we show on an example that this model allows faster learning and better convergence.

[1]  Eric A. Zilli,et al.  Modeling the role of working memory and episodic memory in behavioral tasks , 2008, Hippocampus.

[2]  Alain Dutech,et al.  Apprentissage par renforcement pour les processus décisionnels de Markov partiellement observés Apprendre une extension sélective du passé , 2003, Rev. d'Intelligence Artif..

[3]  Manuela M. Veloso,et al.  Learning of coordination: exploiting sparse interactions in multiagent systems , 2009, AAMAS.

[4]  Jacques Ferber,et al.  Les Systèmes multi-agents: vers une intelligence collective , 1995 .

[5]  Nicolas Sabouret,et al.  Apprentissage par renforcement d'actes de communication dans un système multi-agent , 2010, Rev. d'Intelligence Artif..

[6]  Jonathan D. Cohen,et al.  Learning to Use Working Memory in Partially Observable Environments through Dopaminergic Reinforcement , 2008, NIPS.

[7]  P. Lanzi,et al.  Adaptive Agents with Reinforcement Learning and Internal Memory , 2000 .

[8]  Chris Watkins,et al.  Learning from delayed rewards , 1989 .

[9]  Abdel-Illah Mouaddib,et al.  Collective Decision-Theoretic Planning for Planet Exploration , 2011, 2011 IEEE 23rd International Conference on Tools with Artificial Intelligence.

[10]  Victor R. Lesser,et al.  Communication decisions in multi-agent cooperation: model and experiments , 2001, AGENTS '01.

[11]  Andrew McCallum,et al.  Reinforcement learning with selective perception and hidden state , 1996 .

[12]  A. Kamiya,et al.  Learning of communication codes in multi-agent reinforcement learning problem , 2008, 2008 IEEE Conference on Soft Computing in Industrial Applications.

[13]  John E. Laird,et al.  Extending Cognitive Architecture with Episodic Memory , 2007, AAAI.