# Offline Reinforcement Learning on D4RL HalfCheetah medium-expert v2 Source: [Offline Reinforcement Learning on D4RL HalfCheetah medium-expert v2]() Collection index: [llms.txt]() Data updated: 2026-06-02T08:50:31.575Z Evaluation of Offline Reinforcement Learning performance on the D4RL HalfCheetah environment using the medium-expert v2 dataset configuration, measured by Average True Return. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Offline Reinforcement Learning | | Dataset (including any stated split) | D4RL HalfCheetah medium-expert v2 | | Metric | Avg True Return | | Unit | SCALAR | | Direction | Higher is better | | Results loaded | 17 of 17 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Avg True Return](): Higher is better; unit: SCALAR. ## Recorded results | Method | Avg True Return | Result date | Report paper | | --- | --- | --- | --- | | Optimal Policy | 10364.36 | Not recorded | 68fa6e5b1e209131cea3f53f | | SSR-RRS | 9681.78 | Not recorded | 68fa6e5b1e209131cea3f53f | | SSR-RRS | 9681.78 | Not recorded | 68fa6e5b1e209131cea3f53f | | BVFT | 8799.66 | Not recorded | 68fa6e5b1e209131cea3f53f | | BVFT | 8118.84 | Not recorded | 68fa6e5b1e209131cea3f53f | | CV | 8118.84 | Not recorded | 68fa6e5b1e209131cea3f53f | | CV | 8118.84 | Not recorded | 68fa6e5b1e209131cea3f53f | | PhyB | 109.4 | Not recorded | 6a1e63c9dbd7150506f853f2 | | PMDB | 108.5 | Not recorded | 6a1e63c9dbd7150506f853f2 | | ADM | 103.7 | Not recorded | 6a1e63c9dbd7150506f853f2 | | TD3+BC | 97.9 | Not recorded | 6a1e63c9dbd7150506f853f2 | | FQL | 97.5 | Not recorded | 6a1e63c9dbd7150506f853f2 | | EPQ | 95.7 | Not recorded | 6a1e63c9dbd7150506f853f2 | | RAMBO | 95.4 | Not recorded | 6a1e63c9dbd7150506f853f2 | | CQL | 95 | Not recorded | 6a1e63c9dbd7150506f853f2 | | DMG | 91.1 | Not recorded | 6a1e63c9dbd7150506f853f2 | | MOReL | 80.4 | Not recorded | 6a1e63c9dbd7150506f853f2 | ### Optimal Policy Avg True Return: 10364.36. Result ID: 69902dd13a6c314849feb043. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### SSR-RRS Avg True Return: 9681.78. Result ID: 69902dd13a6c314849feb041. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"splits":"2","re-trained on full dataset":"true"}. This record does not assert independent reproduction. ### SSR-RRS Avg True Return: 9681.78. Result ID: 69902dd13a6c314849feb042. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"splits":"5","re-trained on full dataset":"true"}. This record does not assert independent reproduction. ### BVFT Avg True Return: 8799.66. Result ID: 69902dd13a6c314849feb03e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"variant":"pi + FQE","re-trained on full dataset":"true"}. This record does not assert independent reproduction. ### BVFT Avg True Return: 8118.84. Result ID: 69902dd13a6c314849feb03d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"variant":"pi x FQE","re-trained on full dataset":"true"}. This record does not assert independent reproduction. ### CV Avg True Return: 8118.84. Result ID: 69902dd13a6c314849feb03f. [Reporting paper]() Recorded evaluation configuration: {"splits":"2","re-trained on full dataset":"true"}. This record does not assert independent reproduction. ### CV Avg True Return: 8118.84. Result ID: 69902dd13a6c314849feb040. [Reporting paper]() Recorded evaluation configuration: {"splits":"5","re-trained on full dataset":"true"}. This record does not assert independent reproduction. ### PhyB Avg True Return: 109.4. Result ID: 6a1e9948fbc373592c18daa0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model category":"Model-based"}. This record does not assert independent reproduction. ### PMDB Avg True Return: 108.5. Result ID: 6a1e9948fbc373592c18da9e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model category":"Model-based"}. This record does not assert independent reproduction. ### ADM Avg True Return: 103.7. Result ID: 6a1e9948fbc373592c18da9f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model category":"Model-based"}. This record does not assert independent reproduction. ### TD3+BC Avg True Return: 97.9. Result ID: 6a1e9948fbc373592c18da9a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model category":"Model-free"}. This record does not assert independent reproduction. ### FQL Avg True Return: 97.5. Result ID: 6a1e9948fbc373592c18da99. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model category":"Model-free"}. This record does not assert independent reproduction. ### EPQ Avg True Return: 95.7. Result ID: 6a1e9948fbc373592c18da97. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model category":"Model-free"}. This record does not assert independent reproduction. ### RAMBO Avg True Return: 95.4. Result ID: 6a1e9948fbc373592c18da9d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model category":"Model-based"}. This record does not assert independent reproduction. ### CQL Avg True Return: 95. Result ID: 6a1e9948fbc373592c18da98. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model category":"Model-free"}. This record does not assert independent reproduction. ### DMG Avg True Return: 91.1. Result ID: 6a1e9948fbc373592c18da9b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model category":"Model-free"}. This record does not assert independent reproduction. ### MOReL Avg True Return: 80.4. Result ID: 6a1e9948fbc373592c18da9c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model category":"Model-based"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.