Research map / Reinforcement learning / Policy gradient and actor-critic
Policy Poisoning in Batch Reinforcement Learning and Control
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| poison rewards via convex attack optimization to force target policy | TCE learner · grid world single terminal | attack cost vs clean reward norm: 2.64 vs 21.61 | Helped |
| poison rewards via convex attack optimization | TCE learner · grid world two terminals | attack cost vs clean reward norm: 0.38 vs 11.09 | Helped |
| poison rewards via attack optimization | LQR learner · vehicle LQR | attack cost vs clean reward norm: 0.73 vs 112.94 | Helped |
| apply reward shaping with optimal value potential instead of policy poisoning (ablation) | TCE learner · two-state MDP | learned policy changed | No clear effect |
| use alpha=1 attack cost norm instead of alpha=2 (ablation) | policy poisoning · TCE and LQR | sparsity of attack | Helped |
Closest papers by meaning
- Data Poisoning to Fake a Nash Equilibria for Markov Games (AAAI 2024 · similarity 0.89)
- Optimal Attack and Defense for Reinforcement Learning (AAAI 2024 · similarity 0.88)
- Multi-level Certified Defense Against Poisoning Attacks in Offline Reinforcement Learning (ICLR 2025 · similarity 0.86)
- PolicyGuard: Towards Test-time and Step-level Adversary Defense for Reinforcement Learning Agent (ICML 2026 · similarity 0.86)
- Correlated Proxies: A New Definition and Improved Mitigation for Reward Hacking (ICLR 2025 · similarity 0.86)
- BadRL: Sparse Targeted Backdoor Attack against Reinforcement Learning (AAAI 2024 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.