Research map / Reinforcement learning / Offline RL
Design from Policies: Conservative Test-Time Adaptation for Offline Policy Optimization
Jinxin Liu, Hongyin Zhang, Zifeng Zhuang, Yachen Kang, Donglin Wang, Bin Wang
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 |
|---|---|---|---|
| Use state-adaptive embedding inference instead of keeping the inferred embedding fixed for the episode (ablation) | DROP · HalfCheetah-medium-expert | normalized return | Improved |
| Use gradient-based embedding optimization instead of selecting only among behavior-policy embeddings (ablation) | DROP · HalfCheetah-medium-expert | normalized return | Improved |
| Remove conservative regularization from the score model (ablation) | DROP-Grad and DROP-Grad-Ada · HalfCheetah-medium-expert | normalized return | Hurt |
| Use Rank return-based task decomposition rather than Random or Quantization decomposition (ablation) | DROP · D4RL tasks | normalized return | Improved |
| Replace return-based decomposition with CVAE-based embedding learning (ablation) | DROP · D4RL tasks | normalized return | Improved |
| Use checkpoint-level online fine-tuning (ablation) | DROP · Most evaluated tasks; AntMaze-v0 is an exception | normalized return | Mixed |
| Increase the interval between test-time embedding inferences (ablation) | DROP · AntMaze tasks | normalized return | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- LAPO — LAPO: Latent-Variable Advantage-Weighted Policy Optimization for Offline Reinforcement Learning (NeurIPS 2022) · better on 4 of 6 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Is Value Learning Really the Main Bottleneck in Offline RL? (NeurIPS 2024 · similarity 0.90)
- Model Selection for Off-policy Evaluation: New Algorithms and Experimental Protocol (NeurIPS 2025 · similarity 0.89)
- Offline Data Enhanced On-Policy Policy Gradient with Provable Guarantees (ICLR 2024 · similarity 0.89)
- Adaptive Advantage-Guided Policy Regularization for Offline Reinforcement Learning (ICML 2024 · similarity 0.89)
- Iteratively Refined Behavior Regularization for Offline Reinforcement Learning (NeurIPS 2024 · similarity 0.89)
- Efficient Policy Evaluation with Offline Data Informed Behavior Policy Design (ICML 2024 · similarity 0.88)
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.
