mlsanityData to 14 Oct 2026

Reinforcement Learning with Perturbed Rewards

arXiv 1810.01032 ↗First posted 2 Oct 2018cs.LG cs.CR cs.CVcited by 79

Not on the map. Why?

Cited by 23 in Deep reinforcement learning methods, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

Recent studies have shown that reinforcement learning (RL) models are vulnerable in various noisy scenarios. For instance, the observed reward channel is often subject to noise in practice (e.g., when rewards are collected through sensors), and is therefore not credible. In addition, for applications such as robotics, a deep reinforcement learning (DRL) algorithm can be manipulated to produce arbitrary errors by receiving corrupted rewards. In this paper, we consider noisy RL problems with perturbed rewards, which can be approximated with a confusion matrix. We develop a robust RL framework that enables agents to learn in noisy environments where only perturbed rewards are observed. Our solution framework builds on existing RL/DRL algorithms and firstly addresses the biased noisy reward setting without any assumptions on the true distribution (e.g., zero-mean Gaussian noise as made in previous works). The core ideas of our solution include estimating a reward confusion matrix and defining a set of unbiased surrogate rewards. We prove the convergence and sample complexity of our approach. Extensive experiments on different DRL platforms show that trained policies based on our estimated surrogate reward can achieve higher expected rewards, and converge faster than existing baselines. For instance, the state-of-the-art PPO algorithm is able to obtain 84.6% and 80.8% improvements on average score for five Atari games, with error rates as 10% and 30% respectively.

Key facts. Cited by 79 papers in the map (data through 14 Oct 2026). First posted on arXiv 2 Oct 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2020: 1; 2021: 8; 2022: 10; 2023: 16; 2024: 14; 2025: 18; 2026: 12.

79 papers.

Sort:

By area:

  1. 2025TTRL: Test-Time Reinforcement Learningcited by 275
  2. 2022Wild Patterns Reloaded: A Survey of Machine Learning Security against Training Data Poisoningcited by 94
  3. 2024RoboGSim: A Real2Sim2Real Robotic Gaussian Splatting Simulatorcited by 53
  4. 2025Reinforcement Learning with Verifiable yet Noisy Rewards under Imperfect Verifierscited by 45
  5. 2022On the Robustness of Safe Reinforcement Learning under Observational Perturbationscited by 38
  6. 2024Comprehensive Overview of Reward Engineering and Shaping in Advancing Reinforcement Learning Applicationscited by 38
  7. 2022Trustworthy Reinforcement Learning Against Intrinsic Vulnerabilities: Robustness, Safety, and Generalizabilitycited by 36
  8. 2024Reward-Robust RLHF in LLMscited by 32
  9. 2025RobustVLA: On Robustness of Vision-Language-Action Model against Multi-Modal Perturbationscited by 25
  10. 2023Towards Robust Offline Reinforcement Learning under Diverse Data Corruptioncited by 24
  11. 2021Reinforcement Learning for Feedback-Enabled Cyber Resiliencecited by 19
  12. 2021No-Regret Reinforcement Learning with Heavy-Tailed Rewardscited by 14
38 more
  1. 2023Internally Rewarded Reinforcement Learningcited by 14
  2. 2026An Imperfect Verifier is Good Enough: Learning with Noisy Rewardscited by 13
  3. 2020Offline Reinforcement Learning from Human Feedback in Real-World Sequence-to-Sequence Taskscited by 12
  4. 2021Robust Reinforcement Learning Under Minimax Regret for Green Securitycited by 12
  5. 2023Safe Reinforcement Learning with Dual Robustnesscited by 10
  6. 2023Adv3D: Generating Safety-Critical 3D Objects through Closed-Loop Simulationcited by 10
  7. 2024Large Language Models as Agents in Two-Player Gamescited by 10
  8. 2023Solving Non-Rectangular Reward-Robust MDPs via Frequency Regularizationcited by 8
  9. 2023Improve Robustness of Reinforcement Learning against Observation Perturbations via l_∞ Lipschitz Policy Networkscited by 7
  10. 2025Reinforcement Learning with Stochastic Reward Machinescited by 7
  11. 2022Distributional Reward Estimation for Effective Multi-Agent Deep Reinforcement Learningcited by 6
  12. 2023Data Optimization in Deep Learning: A Surveycited by 6
  13. 2025Wolfpack Adversarial Attack for Robust Multi-Agent Reinforcement Learningcited by 6
  14. 2021Disturbing Reinforcement Learning Agents with Corrupted Rewardscited by 4
  15. 2023Online Estimation and Inference for Robust Policy Evaluation in Reinforcement Learningcited by 4
  16. 2023Noise Distribution Decomposition based Multi-Agent Distributional Reinforcement Learningcited by 4
  17. 2025CrystalGym: A New Benchmark for Materials Discovery Using Reinforcement Learningcited by 4
  18. 2021Improving Robustness of Deep Reinforcement Learning Agents: Environment Attack based on the Critic Networkcited by 3
  19. 2023Connected Superlevel Set in (Deep) Reinforcement Learning and its Application to Minimax Theoremscited by 3
  20. 2023Beyond Expected Return: Accounting for Policy Reproducibility when Evaluating Reinforcement Learning Algorithmscited by 3
  21. 2025The Policy Cliff: A Theoretical Analysis of Reward-Policy Maps in Large Language Modelscited by 3
  22. 2025Robust Policy Expansion for Offline-to-Online RL under Diverse Data Corruptioncited by 3
  23. 2021On Assessing The Safety of Reinforcement Learning algorithms Using Formal Methodscited by 2
  24. 2022Review of Metrics to Measure the Stability, Robustness and Resilience of Reinforcement Learningcited by 2
  25. 2022Deep Reinforcement Learning with Vector Quantized Encodingcited by 2
  26. 2022Security of Deep Reinforcement Learning for Autonomous Driving: A Surveycited by 2
  27. 2023Facilitating Sim-to-real by Intrinsic Stochasticity of Real-Time Simulation in Reinforcement Learning for Robot Manipulationcited by 2
  28. 2024SInViG: A Self-Evolving Interactive Visual Agent for Human-Robot Interactioncited by 2
  29. 2024Strategizing against Q-learners: A Control-theoretical Approachcited by 2
  30. 2024Symmetric Reinforcement Learning Loss for Robust Learning on Diverse Tasks and Model Scalescited by 2
  31. 2025PANDORA: Diffusion Policy Learning for Dexterous Robotic Piano Playingcited by 2
  32. 2025Enhancing Security in Deep Reinforcement Learning: A Comprehensive Survey on Adversarial Attacks and Defensescited by 2
  33. 2026Objective Decoupling in Social Reinforcement Learning: Recovering Ground Truth from Sycophantic Majoritiescited by 2
  34. 2022Reinforcement Learning for Linear Quadratic Control is Vulnerable Under Cost Manipulationcited by 1
  35. 2022Distributed Multi-Agent Deep Reinforcement Learning for Robust Coordination against Noisecited by 1
  36. 2023Policy Resilience to Environment Poisoning Attacks on Reinforcement Learningcited by 1
  37. 2023Exploring Parity Challenges in Reinforcement Learning through Curriculum Learning with Noisy Labelscited by 1
  38. 2024Adapting Image-based RL Policies via Predicted Rewardscited by 1

50 most cited of 79 All 79 (JSON)