Diversified strategies for mitigating adversarial attacks in multiagent systems
Journal
Proceedings of the International Joint Conference on Autonomous Agents and Multiagent Systems, AAMAS
Journal Volume
1
Pages
407-415
Date Issued
2018
Author(s)
Abstract
In this work we consider online decision-making in settings where players want to guard against possible adversarial attacks or other catastrophic failures. To address this, we propose a solution concept in which players have an additional constraint that at each time step they must play a diversified mixed strategy: one that does not put too much weight on any one action. This constraint is motivated by applications such as finance, routing, and resource allocation, where one would like to limit one's exposure to adversarial or catastrophic events while still perfonning well in typical cases. We explore prope rties of diversified strategies in both zero-sum and general-sum games, and provide algorithms for minimizing regret within the family of diversified strategies as well as methods for using taxes or fees to guide standard regret-minimizing players towards divers ified strategies. We also analyze equilibria produced by diversified strategies in general-sum games. We show that surprisingly, requiri ng diversification can actually lead to higher-welfare equilibria, and give strong guarantees on both price of anarchy and the social welfare produced by regret-minimizing diversified agents. We addit ionally give algorithms for finding optimal diversified strategies in distributed settings where one must limit communication overhead. © 2018 International Foundation for Autonomous Agents and Multiagent Systems.
Subjects
Adversarial multiagent systems; Diversified strategies; Game theory; General-sum games; Regret minimization; Risk mitigation
Other Subjects
Decision making; Game theory; Multi agent systems; Catastrophic event; Catastrophic failures; Communication overheads; Diversified strategies; General sum games; On-line decision makings; Regret minimization; Risk mitigation; Autonomous agents
Type
conference paper