mlsanityData to 14 Oct 2026

Reward Poisoning Attacks on Offline Multi-Agent Reinforcement Learning

arXiv 2206.01888 ↗First posted 4 Jun 2022cs.LG cs.AI cs.CRcited by 20

On the map: Offline RL and imitation learning from data → Offline RL algorithms

Abstract (arXiv)

In offline multi-agent reinforcement learning (MARL), agents estimate policies from a given dataset. We study reward-poisoning attacks in this setting where an exogenous attacker modifies the rewards in the dataset before the agents see the dataset. The attacker wants to guide each agent into a nefarious target policy while minimizing the Lp norm of the reward modification. Unlike attacks on single-agent RL, we show that the attacker can install the target policy as a Markov Perfect Dominant Strategy Equilibrium (MPDSE), which rational agents are guaranteed to follow. This attack can be significantly cheaper than separate single-agent attacks. We show that the attack works on various MARL agents including uncertainty-aware learners, and we exhibit linear programs to efficiently solve the attack problem. We also study the relationship between the structure of the datasets and the minimal attack cost. Our work paves the way for studying defense in offline MARL.

Key facts. Cited by 20 papers in the map (data through 14 Oct 2026). Placed in Offline RL algorithms, one of 13 directions in Offline RL and imitation learning from data. Builds on 31 of its 47 references that are in the map. First posted on arXiv 4 Jun 2022.

Builds on

31 of 47 references.

Closest ancestors · 4

Same direction, newest first

  1. 2022COPA: Certifying Robust Policies for Offline Reinforcement Learning against Poisoning Attacks16 Mar
  2. 2021Corruption-Robust Offline Reinforcement Learning11 Jun
  3. 2020Is Pessimism Provably Efficient for Offline RL?30 Dec
  4. 2019Policy Poisoning in Batch Reinforcement Learning and Control13 Oct

Other references · 27

  1. 2019Multi-Agent Reinforcement Learning: A Selective Overview of Theories and Algorithmscited by 567
  2. 2019Adversarial Policies: Attacking Deep Reinforcement Learningcited by 214
  3. 2020Learning Zero-Sum Simultaneous-Move Markov Games Using Function Approximation and Correlated Equilibriumcited by 100
  4. 2020Adaptive Reward-Poisoning Attacks against Reinforcement Learningcited by 86
  5. 2020Policy Teaching via Environment Poisoning: Training-time Adversarial Attacks against Reinforcement Learningcited by 81
  6. 2019Corruption-robust exploration in episodic reinforcement learningcited by 72
  7. 2020Stochastic Linear Bandits Robust to Adversarial Attackscited by 65
  8. 2021Plan Better Amid Conservatism: Offline Multi-Agent Reinforcement Learning with Actor Rectificationcited by 61
  9. 2018Adversarial Attacks on Stochastic Banditscited by 57
  10. 2019Data Poisoning Attacks on Stochastic Banditscited by 53
  11. 2019Deceptive Reinforcement Learning Under Adversarial Manipulations on Cost Signalscited by 50
  12. 2021A Model Selection Approach for Corruption Robust Reinforcement Learningcited by 45
15 more
  1. 2022Pessimistic Minimax Value Iteration: Provably Efficient Equilibrium Learning from Offline Datasetscited by 44
  2. 2020Vulnerability-Aware Poisoning Mechanism for Online RL with Unknown Dynamicscited by 38
  3. 2020Adversarial Attacks on Linear Contextual Banditscited by 34
  4. 2021Offline Decentralized Multi-Agent Reinforcement Learningcited by 33
  5. 2021Robust Policy Gradient against Strong Data Corruptioncited by 30
  6. 2021Provably Efficient Black-Box Action Poisoning Attacks Against Reinforcement Learningcited by 29
  7. 2022When is Offline Two-Player Zero-Sum Markov Game Solvable?cited by 28
  8. 2021Reward Poisoning in Reinforcement Learning: Attacks Against Unknown Learners in Unknown Environmentscited by 26
  9. 2020Policy Teaching in Reinforcement Learning via Environment Poisoning Attackscited by 24
  10. 2018Data Poisoning Attacks in Contextual Banditscited by 23
  11. 2021Defense Against Reward Poisoning Attacks in Reinforcement Learningcited by 21
  12. 2020Robust Stochastic Bandit Algorithms under Probabilistic Unbounded Adversarial Attackcited by 13
  13. 2022Admissible Policy Teaching through Reward Designcited by 12
  14. 2021Saving Stochastic Bandits from Poisoning Attacks via Limited Data Verificationcited by 9
  15. 2021Game Redesign in No-regret Game Playingcited by 8

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 4; 2024: 6; 2025: 7; 2026: 3.

20 papers.

Sort:

By area:

  1. 2023Implicit Poisoning Attacks in Two-Agent Reinforcement Learning: Adversarial Policies for Training-Time Attackscited by 14
  2. 2023Efficient Adversarial Attacks on Online Multi-agent Reinforcement Learningcited by 10
  3. 2024TrajDeleter: Enabling Trajectory Forgetting in Offline Reinforcement Learning Agentscited by 7
  4. 2023Multi-Agent Reinforcement Learning: Methods, Applications, Visionary Prospects, and Challengescited by 5
  5. 2024Corruption-Robust Offline Two-Player Zero-Sum Markov Gamescited by 4
  6. 2025Temporal Logic-Based Multi-Vehicle Backdoor Attacks against Offline RL Agents in End-to-end Autonomous Drivingcited by 4
  7. 2023Minimally Modifying a Markov Game to Achieve Any Nash Equilibrium and Valuecited by 3
  8. 2024On the Hardness of Decentralized Multi-Agent Policy Evaluation under Byzantine Attackscited by 3
  9. 2025Multi-level Certified Defense Against Poisoning Attacks in Offline Reinforcement Learningcited by 3
  10. 2025TrojanTO: Action-Level Backdoor Attacks against Trajectory Optimization Modelscited by 3
  11. 2024Online Poisoning Attack Against Reinforcement Learning under Black-box Environmentscited by 2
  12. 2025Enhancing Security in Deep Reinforcement Learning: A Comprehensive Survey on Adversarial Attacks and Defensescited by 2
8 more
  1. 2024Inception: Efficiently Computable Misinformation Attacks on Markov Gamescited by 1
  2. 2024Provably Efficient Action-Manipulation Attack Against Continuous Reinforcement Learningcited by 1
  3. 2026Efficient Preference Poisoning Attack on Offline RLHFcited by 1
  4. 2025Optimally Installing Strict Equilibriacited by 0
  5. 2025Robust In-Context Reinforcement Learning Under Reward Poisoning Attackscited by 0
  6. 2025Neutral Agent-based Adversarial Policy Learning against Deep Reinforcement Learning in Multi-party Open Systemscited by 0
  7. 2026Corruption-robust Offline Multi-agent Reinforcement Learning From Human Feedbackcited by 0
  8. 2026When Can You Poison Rewards? A Tight Characterization of Reward Poisoning in Linear MDPscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2023Data Poisoning to Fake a Nash Equilibrium in Markov Games
  2. 2025Collapsing Sequence-Level Data-Policy Coverage via Poisoning Attack in Offline Reinforcement Learning
  3. 2025Optimal Perturbation Budget Allocation for Data Poisoning in Offline Reinforcement Learning
  4. 2024Universal Black-Box Reward Poisoning Attack against Offline Reinforcement Learning
  5. 2024Exploiting Structure in Offline Multi-Agent RL: The Benefits of Low Interaction Rank
  6. 2023Offline Learning in Markov Games with General Function Approximation
  7. 2022Execute Order 66: Targeted Data Poisoning for Reinforcement Learning
  8. 2022BAFFLE: Hiding Backdoors in Offline Reinforcement Learning Datasets