A Hybrid Interaction Model for Multi-Agent Reinforcement Learning

Douglas M. Guisi, Richardson Ribeiro, Marcelo Teixeira, André P. Borges, Eden R. Dosciatti, Fabrício Enembreck

2016

Abstract

The main contribution of this paper is to implement a hybrid method of coordination from the combination of interaction models developed previously. The interaction models are based on the sharing of rewards for learning with multiple agents in order to discover interactively good quality policies. Exchange of rewards among agents, when not occur properly, can cause delays in learning or even cause unexpected behavior, making the cooperation inefficient and converging to a non-satisfactory policy. From these concepts, the hybrid method uses the characteristics of each model, reducing possible conflicts between different policy actions with rewards, improving the coordination of agents in reinforcement learning problems. Experimental results show that the hybrid method can accelerate the convergence, rapidly gaining optimal policies even in large spaces of states, exceeding the results of classical approaches to reinforcement learning.

Paper Citation

in Harvard Style

Guisi D., Ribeiro R., Teixeira M., Borges A., Dosciatti E. and Enembreck F. (2016). A Hybrid Interaction Model for Multi-Agent Reinforcement Learning . In Proceedings of the 18th International Conference on Enterprise Information Systems - Volume 1: ICEIS, ISBN 978-989-758-187-8, pages 54-61. DOI: 10.5220/0005832300540061

in Bibtex Style

@conference{iceis16,
author={Douglas M. Guisi and Richardson Ribeiro and Marcelo Teixeira and André P. Borges and Eden R. Dosciatti and Fabrício Enembreck},
title={A Hybrid Interaction Model for Multi-Agent Reinforcement Learning},
booktitle={Proceedings of the 18th International Conference on Enterprise Information Systems - Volume 1: ICEIS,},
year={2016},
pages={54-61},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0005832300540061},
isbn={978-989-758-187-8},
}

in EndNote Style

TY - CONF
JO - Proceedings of the 18th International Conference on Enterprise Information Systems - Volume 1: ICEIS,
TI - A Hybrid Interaction Model for Multi-Agent Reinforcement Learning
SN - 978-989-758-187-8
AU - Guisi D.
AU - Ribeiro R.
AU - Teixeira M.
AU - Borges A.
AU - Dosciatti E.
AU - Enembreck F.
PY - 2016
SP - 54
EP - 61
DO - 10.5220/0005832300540061