Discrete-time mean field control with environment states

Research output: Chapter in book/report/conference proceedingConference contributionResearchpeer review

Authors

  • Anam Tahir

External Research Organisations

  • Technische Universität Darmstadt
View graph of relations

Details

Original languageEnglish
Title of host publication60th IEEE Conference on Decision and Control, CDC 2021
Pages5239 - 5246
Number of pages8
ISBN (electronic)9781665436595
Publication statusPublished - 14 Dec 2021
Externally publishedYes

Publication series

NameProceedings of the IEEE Conference on Decision and Control
Volume2021-December
ISSN (Print)0743-1546
ISSN (electronic)2576-2370

Abstract

Multi-agent reinforcement learning methods have shown remarkable potential in solving complex multi-agent problems but mostly lack theoretical guarantees. Recently, mean field control and mean field games have been established as a tractable solution for large-scale multi-agent problems with many agents. In this work, driven by a motivating scheduling problem, we consider a discrete-time mean field control model with common environment states. We rigorously establish approximate optimality as the number of agents grows in the finite agent case and find that a dynamic programming principle holds, resulting in the existence of an optimal stationary policy. As exact solutions are difficult in general due to the resulting continuous action space of the limiting mean field Markov decision process, we apply established deep reinforcement learning methods to solve the associated mean field control problem. The performance of the learned mean field control policy is compared to typical multi-agent reinforcement learning approaches and is found to converge to the mean field performance for sufficiently many agents, verifying the obtained theoretical results and reaching competitive solutions.

ASJC Scopus subject areas

Cite this

Discrete-time mean field control with environment states. / Tahir, Anam.
60th IEEE Conference on Decision and Control, CDC 2021. 2021. p. 5239 - 5246 (Proceedings of the IEEE Conference on Decision and Control; Vol. 2021-December).

Research output: Chapter in book/report/conference proceedingConference contributionResearchpeer review

Tahir, A 2021, Discrete-time mean field control with environment states. in 60th IEEE Conference on Decision and Control, CDC 2021. Proceedings of the IEEE Conference on Decision and Control, vol. 2021-December, pp. 5239 - 5246. https://doi.org/10.1109/CDC45484.2021.9683749
Tahir, A. (2021). Discrete-time mean field control with environment states. In 60th IEEE Conference on Decision and Control, CDC 2021 (pp. 5239 - 5246). (Proceedings of the IEEE Conference on Decision and Control; Vol. 2021-December). https://doi.org/10.1109/CDC45484.2021.9683749
Tahir A. Discrete-time mean field control with environment states. In 60th IEEE Conference on Decision and Control, CDC 2021. 2021. p. 5239 - 5246. (Proceedings of the IEEE Conference on Decision and Control). doi: 10.1109/CDC45484.2021.9683749
Tahir, Anam. / Discrete-time mean field control with environment states. 60th IEEE Conference on Decision and Control, CDC 2021. 2021. pp. 5239 - 5246 (Proceedings of the IEEE Conference on Decision and Control).
Download
@inproceedings{528fce98c40644cf9011893cd6ec59b5,
title = "Discrete-time mean field control with environment states",
abstract = "Multi-agent reinforcement learning methods have shown remarkable potential in solving complex multi-agent problems but mostly lack theoretical guarantees. Recently, mean field control and mean field games have been established as a tractable solution for large-scale multi-agent problems with many agents. In this work, driven by a motivating scheduling problem, we consider a discrete-time mean field control model with common environment states. We rigorously establish approximate optimality as the number of agents grows in the finite agent case and find that a dynamic programming principle holds, resulting in the existence of an optimal stationary policy. As exact solutions are difficult in general due to the resulting continuous action space of the limiting mean field Markov decision process, we apply established deep reinforcement learning methods to solve the associated mean field control problem. The performance of the learned mean field control policy is compared to typical multi-agent reinforcement learning approaches and is found to converge to the mean field performance for sufficiently many agents, verifying the obtained theoretical results and reaching competitive solutions.",
author = "Anam Tahir",
note = "Publisher Copyright: {\textcopyright} 2021 IEEE.",
year = "2021",
month = dec,
day = "14",
doi = "10.1109/CDC45484.2021.9683749",
language = "English",
series = "Proceedings of the IEEE Conference on Decision and Control",
pages = "5239 -- 5246",
booktitle = "60th IEEE Conference on Decision and Control, CDC 2021",

}

Download

TY - GEN

T1 - Discrete-time mean field control with environment states

AU - Tahir, Anam

N1 - Publisher Copyright: © 2021 IEEE.

PY - 2021/12/14

Y1 - 2021/12/14

N2 - Multi-agent reinforcement learning methods have shown remarkable potential in solving complex multi-agent problems but mostly lack theoretical guarantees. Recently, mean field control and mean field games have been established as a tractable solution for large-scale multi-agent problems with many agents. In this work, driven by a motivating scheduling problem, we consider a discrete-time mean field control model with common environment states. We rigorously establish approximate optimality as the number of agents grows in the finite agent case and find that a dynamic programming principle holds, resulting in the existence of an optimal stationary policy. As exact solutions are difficult in general due to the resulting continuous action space of the limiting mean field Markov decision process, we apply established deep reinforcement learning methods to solve the associated mean field control problem. The performance of the learned mean field control policy is compared to typical multi-agent reinforcement learning approaches and is found to converge to the mean field performance for sufficiently many agents, verifying the obtained theoretical results and reaching competitive solutions.

AB - Multi-agent reinforcement learning methods have shown remarkable potential in solving complex multi-agent problems but mostly lack theoretical guarantees. Recently, mean field control and mean field games have been established as a tractable solution for large-scale multi-agent problems with many agents. In this work, driven by a motivating scheduling problem, we consider a discrete-time mean field control model with common environment states. We rigorously establish approximate optimality as the number of agents grows in the finite agent case and find that a dynamic programming principle holds, resulting in the existence of an optimal stationary policy. As exact solutions are difficult in general due to the resulting continuous action space of the limiting mean field Markov decision process, we apply established deep reinforcement learning methods to solve the associated mean field control problem. The performance of the learned mean field control policy is compared to typical multi-agent reinforcement learning approaches and is found to converge to the mean field performance for sufficiently many agents, verifying the obtained theoretical results and reaching competitive solutions.

UR - http://www.scopus.com/inward/record.url?scp=85126041900&partnerID=8YFLogxK

U2 - 10.1109/CDC45484.2021.9683749

DO - 10.1109/CDC45484.2021.9683749

M3 - Conference contribution

T3 - Proceedings of the IEEE Conference on Decision and Control

SP - 5239

EP - 5246

BT - 60th IEEE Conference on Decision and Control, CDC 2021

ER -