Fault Tolerant Multi-Agent Learning with Adversarial Budget Constraints

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Mguni, David, Sun, Yaqi, Chen, Haojun, Yang, Wanrong, Darabi, Amir, Orimoloye, Larry Olanrewaju, Yang, Yaodong
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866908840165900288
author Mguni, David
Sun, Yaqi
Chen, Haojun
Yang, Wanrong
Darabi, Amir
Orimoloye, Larry Olanrewaju
Yang, Yaodong
author_facet Mguni, David
Sun, Yaqi
Chen, Haojun
Yang, Wanrong
Darabi, Amir
Orimoloye, Larry Olanrewaju
Yang, Yaodong
contents We study robustness to agent malfunctions in cooperative multi-agent reinforcement learning (MARL), a failure mode that is critical in practice yet underexplored in existing theory. We introduce MARTA, a plug-and-play robustness layer that augments standard MARL algorithms with a Switcher-Adversary mechanism which selectively induces malfunctions in performance-critical states. This formulation defines a fault-switching $(N+2)$-player Markov game in which the Switcher chooses when and which agent fails, and the Adversary controls the resulting faulty behaviour via random or worst-case policies. We develop a Q-learning-type scheme and show that the associated Bellman operator is a contraction, yielding existence and uniqueness of the minimax value, convergence to a Markov perfect equilibrium. MARTA integrates seamlessly with MARL algorithms without architectural modification and consistently improves robustness across Traffic Junction (TJ), Level-Based Foraging (LBF), MPE SimpleTag, and SMAC (v2). In these domains, MARTA achieves large gains in final performance of up to 116.7\% in SMAC, 21.4\% in MPE SimpleTag, and 44.6\% in LBF, while significantly reducing failure rates under train-test mismatched fault regimes. These results establish MARTA as a theoretically grounded and practically deployable mechanism for fault-tolerant MARL.
format Preprint
id arxiv_https___arxiv_org_abs_2508_08800
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Fault Tolerant Multi-Agent Learning with Adversarial Budget Constraints
Mguni, David
Sun, Yaqi
Chen, Haojun
Yang, Wanrong
Darabi, Amir
Orimoloye, Larry Olanrewaju
Yang, Yaodong
Multiagent Systems
We study robustness to agent malfunctions in cooperative multi-agent reinforcement learning (MARL), a failure mode that is critical in practice yet underexplored in existing theory. We introduce MARTA, a plug-and-play robustness layer that augments standard MARL algorithms with a Switcher-Adversary mechanism which selectively induces malfunctions in performance-critical states. This formulation defines a fault-switching $(N+2)$-player Markov game in which the Switcher chooses when and which agent fails, and the Adversary controls the resulting faulty behaviour via random or worst-case policies. We develop a Q-learning-type scheme and show that the associated Bellman operator is a contraction, yielding existence and uniqueness of the minimax value, convergence to a Markov perfect equilibrium. MARTA integrates seamlessly with MARL algorithms without architectural modification and consistently improves robustness across Traffic Junction (TJ), Level-Based Foraging (LBF), MPE SimpleTag, and SMAC (v2). In these domains, MARTA achieves large gains in final performance of up to 116.7\% in SMAC, 21.4\% in MPE SimpleTag, and 44.6\% in LBF, while significantly reducing failure rates under train-test mismatched fault regimes. These results establish MARTA as a theoretically grounded and practically deployable mechanism for fault-tolerant MARL.
title Fault Tolerant Multi-Agent Learning with Adversarial Budget Constraints
topic Multiagent Systems
url https://arxiv.org/abs/2508.08800