Research map / Reinforcement learning / Offline RL
Peak-Return Greedy Slicing: Subtrajectory Selection for Transformer-based Offline RL
Zhiwei Xu, Miduo Cui, Dapeng Li, Zhihao Liu, Haifeng Zhang, Hangyu Mao, Guoliang Fan, Bin Zhang
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 |
|---|---|---|---|
| add PRGS subtrajectory slicing to Decision Transformer | DT · D4RL Maze2D (average) | normalized return: 40.9 -> 100.1 | Improved |
| add PRGS to Decision Transformer | DT · D4RL Gym (average) | normalized return: 75.3 -> 82.9 | Improved |
| add PRGS to PDiT | PDiT · D4RL AntMaze (average) | normalized return: 38.7 -> 37.1 | Hurt |
| add PRGS to DT | DT · D4RL Adroit door tasks | normalized return: door-human 22.7 -> 7.8 | Mixed |
| add PRGS to BC | Transformer BC · AuctionNet (average) | score: 269.9 -> 279.0 | Improved |
| add PRGS to PDiT | PDiT · BabyAI (average) | success rate: 63.0 -> 65.9 | Improved |
| add PRGS on narrow-coverage random and expert datasets | PDiT · Gym random/expert (average) | normalized return: 54.0 -> 54.2 | No clear effect |
| add adaptive history truncation at evaluation (ablation) | PRGS · Maze2D | normalized return | Improved |
| use very small particle number n in top-n optimistic estimate (ablation) | PRGS · Maze2D | normalized return | Hurt |
| use very large n (conservative value estimate) (ablation) | PRGS · Maze2D | normalized return | Hurt |
| train on top 10% or 20% of trajectories instead of PRGS slicing (ablation) | BC, DT, PDiT · Maze2D | normalized return | No clear effect |
| use optimistic MMD estimator rather than w/o Optimistic or Q-loss (ablation) | PRGS · hopper-medium and hopper-medium-replay | normalized return | Improved |
Closest papers by meaning
- Reinformer: Max-Return Sequence Modeling for Offline RL (ICML 2024 · similarity 0.90)
- Critic-Guided Decision Transformer for Offline Reinforcement Learning (AAAI 2024 · similarity 0.90)
- In-Dataset Trajectory Return Regularization for Offline Preference-based Reinforcement Learning (AAAI 2025 · similarity 0.89)
- ContraDiff: Planning Towards High Return States via Contrastive Learning (ICLR 2025 · similarity 0.89)
- Offline RL with Observation Histories: Analyzing and Improving Sample Complexity (ICLR 2024 · similarity 0.89)
- In-Context Reinforcement Learning From Suboptimal Historical Data (ICML 2025 · similarity 0.89)
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.
