Networked multi-agent reinforcement learning with emergent communication

Gupta, S and Hazra, R and Dukkipati, A (2020) Networked multi-agent reinforcement learning with emergent communication. In: Proceedings of the International Joint Conference on Autonomous Agents and Multiagent Systems, AAMAS, 19-13 May 2020, Virtual, Auckland, pp. 1858-1860.

PDF Net-Multi-Agent Rei- Lea-Eme.pdf - Published Version Restricted to Registered users only Download (491kB) | Request a copy

Abstract

We develop a Multi-Agent Reinforcement Learning (MARL) method that finds approximately optimal policies for cooperative agents that co-exist in an environment. Central to achieving this is how the agents learn to communicate with each other. Can they together develop a language while learning to perform a common task? We formulate and study a MARL problem where cooperative agents are connected via a fixed underlying network. These agents communicate along the edges of this network by exchanging discrete symbols. However, the semantics of these symbols are not predefined and have to be learned during the training process. We propose a method for training these agents using emergent communication. We demonstrate the applicability of the proposed framework by applying it to the problem of managing traffic controllers, where we achieve state-of-the-art performance (as compared to several strong baselines) and perform a detailed analysis of the emergent communication. © 2020 International Foundation for Autonomous Agents and Multiagent Systems (IFAAMAS). All rights reserved.

Item Type:	Conference Paper
Publication:	Proceedings of the International Joint Conference on Autonomous Agents and Multiagent Systems, AAMAS
Publisher:	International Foundation for Autonomous Agents and Multiagent Systems (IFAAMAS)
Additional Information:	International Foundation for Autonomous Agents and Multiagent Systems (IFAAMAS)The copyright of this article belongs to
Keywords:	Fertilizers; Multi agent systems; Reinforcement learning; Semantics, Cooperative agents; Multi-agent reinforcement learning; Optimal policies; State-of-the-art performance; Traffic controllers; Training process; Underlying networks, Autonomous agents
Department/Centre:	Division of Mechanical Sciences > Materials Engineering (formerly Metallurgy)
Date Deposited:	15 Mar 2021 05:22
Last Modified:	09 Dec 2022 10:02
URI:	https://eprints.iisc.ac.in/id/eprint/67215

Actions (login required)

View Item