Embodied imitation-enhanced reinforcement learning in multi-agent systems

Erbas, M. D., Winfield, A. F. and Bull, L. (2014) Embodied imitation-enhanced reinforcement learning in multi-agent systems. Adaptive Behavior, 22 (1). pp. 31-50. ISSN 1059-7123 Available from: http://eprints.uwe.ac.uk/21571

[img] PDF

Publisher's URL: http://dx.doi.org/10.1177/1059712313500503


Imitation is an example of social learning in which an individual observes and copies another’s actions. This paper presents a new method for using imitation as a way of enhancing the learning speed of individual agents that employ a well-known reinforcement learning algorithm, namely Q-learning. Compared with other research that uses imitation with reinforcement learning, our method uses imitation of purely observed behaviours to enhance learning, with no internal state access or sharing of experiences between agents. The paper evaluates our imitation-enhanced reinforcement learning approach in both simulation and with real robots in continuous space. Both simulation and real robot experimental results show that the learning speed of the group is improved.

Item Type:Article
Uncontrolled Keywords:embodied imitation, reinforcement q-learning, social learning, multi-agent systems
Faculty/Department:Faculty of Environment and Technology
ID Code:21571
Deposited By: Professor A. Winfield
Deposited On:20 Sep 2013 13:35
Last Modified:31 Jan 2017 22:32

Request a change to this item

Total Document Downloads in Past 12 Months

Document Downloads

Total Document Downloads

More statistics for this item...