mlsanityData to 14 Oct 2026

Data Poisoning Attacks on Stochastic Bandits

arXiv 1905.06494 ↗First posted 16 May 2019cs.LG cs.CR stat.MLcited by 53

Not on the map. Why?

Abstract (arXiv)

Stochastic multi-armed bandits form a class of online learning problems that have important applications in online recommendation systems, adaptive medical treatment, and many others. Even though potential attacks against these learning algorithms may hijack their behavior, causing catastrophic loss in real-world applications, little is known about adversarial attacks on bandit algorithms. In this paper, we propose a framework of offline attacks on bandit algorithms and study convex optimization based attacks on several popular bandit algorithms. We show that the attacker can force the bandit algorithm to pull a target arm with high probability by a slight manipulation of the rewards in the data. Then we study a form of online attacks on bandit algorithms and propose an adaptive attack strategy against any bandit algorithm without the knowledge of the bandit algorithm. Our adaptive attack strategy can hijack the behavior of the bandit algorithm to suffer a linear regret with only a logarithmic cost to the attacker. Our results demonstrate a significant security threat to stochastic bandits.

Key facts. Cited by 53 papers in the map (data through 14 Oct 2026). First posted on arXiv 16 May 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2019: 1; 2020: 0; 2021: 18; 2022: 7; 2023: 14; 2024: 6; 2025: 4; 2026: 3.

53 papers.

Sort:

By area:

  1. 2019Policy Poisoning in Batch Reinforcement Learning and Controlcited by 70
  2. 2021Robust Stochastic Linear Contextual Bandits Under Adversarial Attackscited by 33
  3. 2021Provably Efficient Black-Box Action Poisoning Attacks Against Reinforcement Learningcited by 29
  4. 2021Reward Poisoning in Reinforcement Learning: Attacks Against Unknown Learners in Unknown Environmentscited by 26
  5. 2021Defense Against Reward Poisoning Attacks in Reinforcement Learningcited by 21
  6. 2021On Optimal Robustness to Adversarial Corruption in Online Decision Problemscited by 21
  7. 2021Linear Contextual Bandits with Adversarial Corruptionscited by 21
  8. 2022Understanding the Limits of Poisoning Attacks in Episodic Reinforcement Learningcited by 21
  9. 2021Improved Corruption Robust Algorithms for Episodic Reinforcement Learningcited by 20
  10. 2021Exploiting Heterogeneity in Robust Federated Best-Arm Identificationcited by 20
  11. 2022Rank List Sensitivity of Recommender Systems to Interaction Perturbationscited by 20
  12. 2022Reward Poisoning Attacks on Offline Multi-Agent Reinforcement Learningcited by 20
38 more
  1. 2025Joint-GCG: Unified Gradient-Based Poisoning Attacks on Retrieval-Augmented Generation Systemscited by 16
  2. 2021Poisoning Attack against Estimating from Pairwise Comparisonscited by 14
  3. 2024Trustworthy Distributed AI Systems: Robustness, Privacy, and Governancecited by 14
  4. 2023Best-of-Both-Worlds Algorithms for Linear Contextual Banditscited by 11
  5. 2021Accumulative Poisoning Attacks on Real-time Datacited by 10
  6. 2021When Are Linear Stochastic Bandits Attackable?cited by 10
  7. 2023Concentrated Differential Privacy for Banditscited by 10
  8. 2021Saving Stochastic Bandits from Poisoning Attacks via Limited Data Verificationcited by 9
  9. 2022Efficient Reward Poisoning Attacks on Online Deep Reinforcement Learningcited by 9
  10. 2021Robust Bandit Learning with Imperfect Contextcited by 8
  11. 2021Game Redesign in No-regret Game Playingcited by 8
  12. 2021Combinatorial Bandits under Strategic Manipulationscited by 6
  13. 2022Federated Multi-Armed Bandits Under Byzantine Attackscited by 6
  14. 2023Adversarial Attacks on Adversarial Banditscited by 6
  15. 2023Adversarial Attacks on Online Learning to Rank with Click Feedbackcited by 6
  16. 2023Adversarial Attacks on Online Learning to Rank with Stochastic Click Modelscited by 6
  17. 2022Collaborative Linear Bandits with Adversarial Agents: Near-Optimal Regret Boundscited by 5
  18. 2022Reward Delay Attacks on Deep Reinforcement Learningcited by 5
  19. 2023Black-Box Targeted Reward Poisoning Attack Against Online Deep Reinforcement Learningcited by 5
  20. 2021Disturbing Reinforcement Learning Agents with Corrupted Rewardscited by 4
  21. 2021Best-of-All-Worlds Bounds for Online Learning with Feedback Graphscited by 4
  22. 2023Backdoor Learning for NLP: Recent Advances, Challenges, and Future Research Directionscited by 4
  23. 2021Efficient Action Poisoning Attacks on Linear Contextual Banditscited by 3
  24. 2023Online Corrupted User Detection and Regret Minimizationcited by 3
  25. 2023Byzantine-Resilient Decentralized Multi-Armed Banditscited by 3
  26. 2025Practical Adversarial Attacks on Stochastic Bandits via Fake Data Injectioncited by 3
  27. 2023CRIMED: Lower and Upper Bounds on Regret for Bandits with Unbounded Stochastic Corruptioncited by 2
  28. 2023Adversarial Attacks on Combinatorial Multi-Armed Banditscited by 2
  29. 2024Stealthy Adversarial Attacks on Stochastic Multi-Armed Banditscited by 2
  30. 2024Robust Q-Learning under Corrupted Rewardscited by 2
  31. 2024Robust Thompson Sampling Algorithms Against Reward Poisoning Attackscited by 2
  32. 2026Efficient Adversarial Attacks on High-dimensional Offline Banditscited by 2
  33. 2024Stochastic Bandits Robust to Adversarial Attackscited by 1
  34. 2025Corruption-Tolerant Asynchronous Q-Learning with Near-Optimal Ratescited by 1
  35. 2026Efficient Preference Poisoning Attack on Offline RLHFcited by 1
  36. 2023Robust and differentially private stochastic linear banditscited by 0
  37. 2023Reward Teaching for Federated Multi-armed Banditscited by 0
  38. 2023Adversarial Attacks on Cooperative Multi-agent Banditscited by 0

50 most cited of 53 All 53 (JSON)