Logo image
Back
Optimal Cost Constrained Adversarial Attacks For Multiple Agent Systems
Preprint   Open access

Optimal Cost Constrained Adversarial Attacks For Multiple Agent Systems

Ziqing Lu, Guanlin Liu, Lifeng Cai and Weiyu Xu
ArXiv.org
Cornell University
11/01/2023
DOI: 10.48550/arxiv.2311.00859
url
https://doi.org/10.48550/arxiv.2311.00859View
Preprint (Author's original)This preprint has not been evaluated by subject experts through peer review. Preprints may undergo extensive changes and/or become peer-reviewed journal articles. Open Access

Abstract

Finding optimal adversarial attack strategies is an important topic in reinforcement learning and the Markov decision process. Previous studies usually assume one all-knowing coordinator (attacker) for whom attacking different recipient (victim) agents incurs uniform costs. However, in reality, instead of using one limitless central attacker, the attacks often need to be performed by distributed attack agents. We formulate the problem of performing optimal adversarial agent-to-agent attacks using distributed attack agents, in which we impose distinct cost constraints on each different attacker-victim pair. We propose an optimal method integrating within-step static constrained attack-resource allocation optimization and between-step dynamic programming to achieve the optimal adversarial attack in a multi-agent system. Our numerical results show that the proposed attacks can significantly reduce the rewards received by the attacked agents.

Details

Metrics

6 Record Views
Logo image