Robust Event-Driven Interactions in Cooperative Multi-agent Learning

Conference Paper (2022)
Author(s)

D. Jarne Ornia (TU Delft - Team Manuel Mazo Jr)

M Mazo Jr. (TU Delft - Team Manuel Mazo Jr)

Research Group
Team Manuel Mazo Jr
Copyright
© 2022 D. Jarne Ornia, M. Mazo
DOI related publication
https://doi.org/10.1007/978-3-031-15839-1_16
More Info
expand_more
Publication Year
2022
Language
English
Copyright
© 2022 D. Jarne Ornia, M. Mazo
Research Group
Team Manuel Mazo Jr
Bibliographical Note
Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.@en
Pages (from-to)
281-297
ISBN (print)
978-3-031-15838-4
ISBN (electronic)
978-3-031-15839-1
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

We present an approach to safely reduce the communication required between agents in a Multi-Agent Reinforcement Learning system by exploiting the inherent robustness of the underlying Markov Decision Process. We compute robustness certificate functions (off-line), that give agents a conservative indication of how far their state measurements can deviate before they need to update other agents in the system with new measurements. This results in fully distributed decision functions, enabling agents to decide when it is necessary to communicate state variables. We derive bounds on the optimality of the resulting systems in terms of the discounted sum of rewards obtained, and show these bounds are a function of the design parameters. Additionally, we extend the results for the case where the robustness surrogate functions are learned from data, and present experimental results demonstrating a significant reduction in communication events between agents.

Files

978_3_031_15839_1_16.pdf
(pdf | 0.92 Mb)
- Embargo expired in 01-03-2023
License info not available