Deakin University
Browse

File(s) under permanent embargo

Multi-agent Deep Reinforcement Learning with Human Strategies

Version 2 2024-06-06, 07:48
Version 1 2018-07-12, 15:59
conference contribution
posted on 2024-06-06, 07:48 authored by T Nguyen, ND Nguyen, S Nahavandi
Deep learning has enabled traditional reinforcement learning methods to deal with high-dimensional problems. However, one of the disadvantages of deep reinforcement learning methods is the limited exploration capacity of learning agents. In this paper, we introduce an approach that integrates human strategies to increase the exploration capacity of multiple deep reinforcement learning agents. We also report the development of our own multi-agent environment called Multiple Tank Defence to simulate the proposed approach. The results show the significant performance improvement of multiple agents that have learned cooperatively with human strategies. This implies that there is a critical need for human intellect teamed with machines to solve complex problems. In addition, the success of this simulation indicates that our multi-agent environment can be used as a testbed platform to develop and validate other multi-agent control algorithms.

History

Volume

2019-February

Pagination

1357-1362

Location

Melbourne, Vic.

Start date

2019-02-13

End date

2019-02-15

ISSN

2643-2978

ISBN-13

9781538663769

Language

English

Publication classification

E1 Full written paper - refereed

Title of proceedings

ICIT 2019 : Proceedings of the IEEE Industrial Technology 2019 Conference

Event

IEEE Industrial Technology Conference (2019 Melbourne, Vic.)

Publisher

IEEE

Place of publication

Piscataway, N.J.