Research map / Reinforcement learning / Offline RL
Accountability in Offline Reinforcement Learning: Explaining Decisions with a Corpus of Examples
Hao Sun, Alihan Hüyük, Daniel Jarrett, Mihaela van der Schaar
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 |
|---|---|---|---|
| Increase the epsilon quantile from 0.5 to 0.9, filtering fewer candidate actions (ablation) | AOC · Pendulum-Het | performance: -1.25 ± 0.40 -> -503.39 ± 301.60 | Hurt |
| Increase the epsilon quantile from 0.5 to 1.0, filtering fewer candidate actions (ablation) | AOC · Pendulum-Het | performance: -1.25 ± 0.40 -> -853.29 ± 138.43 | Hurt |
| Decrease the epsilon quantile from 0.5 to 0.3 (ablation) | AOC · Pendulum-Het | performance: -1.25 ± 0.40 -> -2.05 ± 0.45 | Hurt |
| Use the corpus-residual filter to exclude actions with larger residuals (ablation) | AOC · Pendulum-Het | cumulative reward: -853.29 ± 138.43 -> -1.25 ± 0.40 | Improved |
| Increase the proportion of offline trajectories from expert π1 | AOC · Continuous Maze | upper-gate route selection | Improved |
| Adjust the proportion of reference behaviors through re-sampling or sub-sampling | AOC · Continuous Maze | success rate | No clear effect |
| Increase the offline dataset size | MFRL · Pendulum-Het | cumulative reward: -4.1 ± 2.76 -> -15.27 ± 6.46 | Hurt |
Closest papers by meaning
- Offline Safe Reinforcement Learning Using Trajectory Classification (AAAI 2025 · similarity 0.86)
- Tackling Data Corruption in Offline Reinforcement Learning via Sequence Modeling (ICLR 2025 · similarity 0.86)
- Simple Ingredients for Offline Reinforcement Learning (ICML 2024 · similarity 0.86)
- When should we prefer Decision Transformers for Offline Reinforcement Learning? (ICLR 2024 · similarity 0.86)
- A Clean Slate for Offline Reinforcement Learning (NeurIPS 2025 · similarity 0.86)
- Adversarially Trained Weighted Actor-Critic for Safe Offline Reinforcement Learning (NeurIPS 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.
