Publication
Title
Model checking for adversarial multi-agent reinforcement learning with reactive defense methods
Author
Abstract
Cooperative multi-agent reinforcement learning (CMARL) enables agents to achieve a common objective. However, the safety (a.k.a. robustness) of the CMARL agents operating in critical environments is not guaranteed. In particular, agents are susceptible to adversarial noise in their observations that can mislead their decision-making. So-called denoisers aim to remove adversarial noise from observations, yet, they are often error-prone. A key challenge for any rigorous safety verification technique in CMARL settings is the large number of states and transitions, which generally prohibits the construction of a (monolithic) model of the whole system. In this paper, we present a verification method for CMARL agents in settings with or without adversarial attacks or denoisers. Our method relies on a tight integration of CMARL and a verification technique referred to as model checking. We showcase the applicability of our method on various benchmarks from different domains. Our experiments show that our method is indeed suited to verify CMARL agents and that it scales better than a naive approach to model checking.
Language
English
Source (journal)
Proceedings of the International Conference on Automated Planning and Scheduling
Source (book)
International Conference on Automated Planning and Scheduling, July 8–13, 2023, Prague, Czech Republic
Publication
2023
ISBN
978-1-57735-881-7
DOI
10.1609/ICAPS.V33I1.27191
Volume/pages
33 :1 (2023) , p. 162-170
Full text (Publisher's DOI)
Full text (open access)
UAntwerpen
Faculty/Department
Research group
Project info
SAILor: Safe Artificial Intelligence and Learning for Verification.
Publication type
Subject
Affiliation
Publications with a UAntwerp address
External links
Record
Identifier
Creation 23.09.2023
Last edited 17.06.2024
To cite this reference