Research map / Reinforcement learning / Offline RL
Offline policy evaluation and selection
Evaluate and select policies from offline data using augmentation and re-weighting.
8 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2021–2024) · an approach in Offline RL, keyword cluster “offline online · transformer · online reinforcement”
State History Tried Preprints Who Export
Where it stands
Not enough benchmark data Fewer than four papers report the same benchmark, so there is no record to follow.
6 papers in 2023–24 and 0 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic shrank, from 2.5% to 0.0%.
How this is decided
A benchmark is tracked when at least four papers of this approach report their own method on the same dataset and metric in at least two different years. Its record is the best number reported up to 2024 compared with the best reported in 2025–26: still improving means at least 1% better, flat means at most 0.2% better with at least two papers from 2025–26 trying. Numbers on a different scale from the rest (a fraction among percentages) and far outliers are left out. Results come from different papers with different setups, so they are signals, not head-to-head comparisons. The share of papers is counted at ICML, NeurIPS, ICLR, CVPR and AAAI.
How it got here
Most cited paper of each year
- 2021 · Benchmarks for Deep Off-Policy Evaluation (ICLR 2021 · 21 citations)
- 2023 · Distributional Offline Policy Evaluation with Predictive Error Guarantees (ICML 2023 · 0 citations)
- 2024 · Probabilistic Offline Policy Ranking with Approximate Bayesian Computation (AAAI 2024 · 2 citations)
Earliest papers here
- Benchmarks for Deep Off-Policy Evaluation (ICLR 2021 · 21 citations)
- Active Offline Policy Selection (NeurIPS 2021 · 0 citations)
- Distributional Offline Policy Evaluation with Predictive Error Guarantees (ICML 2023 · 0 citations)
What has been tried
7 papers here report 67 measured changes (a component added, removed or swapped, compared with a baseline): helped 63%, no clear effect 7%, hurt 22%, mixed 7%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| use POPR-EABC instead of direct action-agreement ranking | MountainCar · NDCG | 0.9829 -> 0.9992 | Helped | Probabilistic Offline Policy Ranking with Approximate Bayesian Computation (AAAI 2024 · 2 citations) |
| Use IS as the centering estimator | Sepsis-MDP N=200 and Sepsis-POMDP N=1000 · MSE | 0.2205 -> 0.2181; 0.2749 -> 0.2720 | Helped | OPERA: Automatic Offline Policy Evaluation with Re-weighted Aggregates of Multiple Estimators (NeurIPS 2024 · 1 citation) |
| Add trajectory augmentation | IE, 6th semester · student performance | Helped | Off-Policy Selection for Initiating Human-Centric Experimental Design (NeurIPS 2024 · 0 citations) |
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 6 papers; from OpenAlex and the papers' affiliations).
- Duke University · 2 papers
- North Carolina State University · 2 papers
Use it
Download BibTeX of 8 key papersOpen the topic in the interactive map
Get alerts. Describe your idea in scime atlas and save it (free scime account): you get a weekly email when new preprints land close to it.
Related approaches
Same keyword cluster
- Offline-to-online fine-tuning (99 papers)
- Diffusion models for offline decision making (82 papers)
- Decision Transformer sequence models (40 papers)
Elsewhere in the topic
- Offline RL sample complexity theory (156 papers)
- Simple practical offline RL (144 papers)
- Pessimism and uncertainty in offline RL (59 papers)
- Behavior regularization for offline RL (35 papers)
Benchmark numbers, ablations and baseline numbers are extracted by AI from each paper's full text and can be wrong; the paper is the authority. Data: results index of 2026-11-02, citation counts from OpenAlex (2026-10-09).