Premium
Training Multiagent Systems by Q‐Learning: Approaches and Empirical Results
Author(s) -
LopezGuede Jose Manuel,
FernandezGau Borja,
Graña Manuel,
Zulueta Ekaitz
Publication year - 2015
Publication title -
computational intelligence
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.353
H-Index - 52
eISSN - 1467-8640
pISSN - 0824-7935
DOI - 10.1111/coin.12035
Subject(s) - curse of dimensionality , reinforcement learning , computer science , artificial intelligence , multi agent system , field (mathematics) , control (management) , machine learning , robot learning , robot , mobile robot , mathematics , pure mathematics
Multiagent systems are increasingly present in computational environments. However, the problem of agent design or control is an open research field. Reinforcement learning approaches offer solutions that allow autonomous learning with minimal supervision. The Q‐learning algorithm is a model‐free reinforcement learning solution that has proven its usefulness in single‐agent domains; however, it suffers from dimensionality curse when applied to multiagent systems. In this article, we discuss two approaches, namely TRQ‐learning and distributed Q‐learning, that overcome the limitations of Q‐learning offering feasible solutions. We test these approaches in two separate domains. The first is the control of a hose by a team of robots. The second is the trash disposal problem. Computational results show the effectiveness of Q‐learning solutions to multiagent systems’ control.