Learning-based Physical Layer Communications for Multiagent Collaboration

Arsham Mostaani, Osvaldo Simeone, Symeon Chatzinotas, Bjorn Ottersten

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Scopus citations

Abstract

Consider a collaborative task carried out by two autonomous agents that can communicate over a noisy channel. Each agent is only aware of its own state, while the accomplishment of the task depends on the value of the joint state of both agents. As an example, both agents must simultaneously reach a certain location of the environment, while only being aware of their own positions. Assuming the presence of feedback in the form of a common reward to the agents, a conventional approach would apply separately: (i) an off-the-shelf coding and decoding scheme in order to enhance the reliability of the communication of the state of one agent to the other; and (ii) a standard multiagent reinforcement learning strategy to learn how to act in the resulting environment. In this work, it is argued that the performance of the collaborative task can be improved if the agents learn how to jointly communicate and act. In particular, numerical results for a baseline grid world example demonstrate that the jointly learned policy carries out compression and unequal error protection by leveraging information about the action policy.

Original languageEnglish (US)
Title of host publication2019 IEEE 30th Annual International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781538681107
DOIs
StatePublished - Sep 2019
Externally publishedYes
Event30th IEEE Annual International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC 2019 - Istanbul, Turkey
Duration: Sep 8 2019Sep 11 2019

Publication series

NameIEEE International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC
Volume2019-September

Conference

Conference30th IEEE Annual International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC 2019
Country/TerritoryTurkey
CityIstanbul
Period9/8/199/11/19

All Science Journal Classification (ASJC) codes

  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Learning-based Physical Layer Communications for Multiagent Collaboration'. Together they form a unique fingerprint.

Cite this