Research map / Reinforcement learning / Policy gradient and actor-critic
Supervised Off-Policy Ranking
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 |
|---|---|---|---|
| Use supervised pairwise ranking of policies with known performance instead of unsupervised return estimation | SOPR-T · D4RL tasks, Test Set I | rank correlation and regret | Mixed |
| Use supervised pairwise ranking of policies with known performance instead of unsupervised return estimation | SOPR-T · D4RL tasks, Test Set II | rank correlation and regret | Helped |
| Replace the Transformer encoder with an MLP encoder (ablation) | SOPR-T · 12 tasks, Test Set I | rank correlation | Hurt |
| Replace the Transformer encoder with an MLP encoder (ablation) | SOPR-T · 12 tasks, Test Set II | rank correlation | Hurt |
| Reduce dataset size (ablation) | SOPR-T · D4RL tasks, 4k–32k data | ranking performance | Hurt |
| Use fewer sampled subsets at inference (ablation) | SOPR-T · D4RL tasks | average ranking performance | No clear effect |
| Use fewer sampled subsets at inference (ablation) | SOPR-T · D4RL tasks | standard deviation of ranking performance | Hurt |
| Train and test on policies with a larger distribution distance | SOPR-T · Test Set I versus Test Set II | ranking performance | Hurt |
Closest papers by meaning
- OPERA: Automatic Offline Policy Evaluation with Re-weighted Aggregates of Multiple Estimators (NeurIPS 2024 · similarity 0.89)
- Probabilistic Offline Policy Ranking with Approximate Bayesian Computation (AAAI 2024 · similarity 0.88)
- Cross-Validated Off-Policy Evaluation (AAAI 2025 · similarity 0.87)
- Cross-Domain Off-Policy Evaluation and Learning for Contextual Bandits (ICLR 2025 · similarity 0.87)
- Abstract Reward Processes: Leveraging State Abstraction for Consistent Off-Policy Evaluation (NeurIPS 2024 · similarity 0.86)
- Off-policy Evaluation Beyond Overlap: Sharp Partial Identification Under Smoothness (ICML 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.