Research map / Reinforcement learning / MDPs and planning
Seeing is not Believing: Robust Reinforcement Learning against Spurious Correlation
Wenhao Ding, Laixi Shi, Yuejie Chi, Ding Zhao
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| add causal graph G_phi (vs removing it) (ablation) | RSC-SAC · Behavior shifted | normalized reward: 0.51 -> 1.02 | Improved |
| add causal graph G_phi (vs removing it) (ablation) | RSC-SAC · Lift shifted | normalized reward: 0.79 -> 0.98 | Improved |
| add transition model P_theta (vs removing it) (ablation) | RSC-SAC · Behavior shifted | normalized reward: 0.41 -> 1.02 | Improved |
| add transition model P_theta (vs removing it) (ablation) | RSC-SAC · Crossing shifted | normalized reward: 0.89 -> 1.04 | Improved |
| add confounder distribution P^c (vs removing it) (ablation) | RSC-SAC · Behavior shifted | normalized reward: 0.66 -> 1.02 | Improved |
| add confounder distribution P^c (vs removing it) (ablation) | RSC-SAC · Lift shifted | normalized reward: 0.90 -> 0.98 | Improved |
| add L1 sparsity regularization to SAC | SAC · distraction correlation tasks | normalized reward | Improved |
| add L1 sparsity regularization to SAC | SAC · composition correlation tasks | normalized reward | No clear effect |
| use traditional robust RL (RMDP-G, RMDP-U, ATLA) outside their uncertainty set | SAC · shifted environments | normalized reward | Hurt |
| RSC-SAC vs RMDP-0.01 under Gaussian noise 0.1 | RSC-SAC · Lift-0.01 | normalized reward: 0.96 vs 0.94 | No clear effect |
| RSC-SAC vs RMDP-0.1 in nominal Lift | RSC-SAC · Lift-0 | normalized reward: 0.85 -> 0.96 | Improved |
| increase augmentation ratio beta to 1% (ablation) | RSC-SAC · nominal and shifted | normalized reward | No clear effect |
Closest papers by meaning
- Towards Robust Offline Reinforcement Learning under Diverse Data Corruption (ICLR 2024 · similarity 0.88)
- Distributionally Robust Reinforcement Learning with Interactive Data Collection: Fundamental Hardness and Near-Optimal Algorithms (NeurIPS 2024 · similarity 0.87)
- Confounding Robust Deep Reinforcement Learning: A Causal Approach (NeurIPS 2025 · similarity 0.87)
- Time-Constrained Robust MDPs (NeurIPS 2024 · similarity 0.86)
- State Entropy Regularization for Robust Reinforcement Learning (NeurIPS 2025 · similarity 0.86)
- Delphic Offline Reinforcement Learning under Nonidentifiable Hidden Confounding (ICLR 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.
