Research map / Architectures and efficiency / Attention and transformers
Elastic Decision Transformer
Yueh-Hua Wu, Xiaolong Wang, Masashi Hamaya
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 |
|---|---|---|---|
| fixed history length 20 instead of variable (ablation) | EDT · D4RL hopper-medium-replay | normalized return: 67.6 -> 89.0 | Improved |
| fixed history length 5 instead of variable (ablation) | EDT · D4RL hopper-medium-replay | normalized return: 62.6 -> 89.0 | Improved |
| fixed history length 20 instead of variable (ablation) | EDT · D4RL walker-medium-replay | normalized return: 64.5 -> 74.8 | Improved |
| fixed history length 5 instead of variable (ablation) | EDT · D4RL walker-medium-replay | normalized return: 44.3 -> 74.8 | Improved |
| fixed history length 20 instead of variable (ablation) | EDT · D4RL halfcheetah-medium | normalized return: 42.0 -> 42.5 | No clear effect |
| fixed history length 20 vs variable, summed over 8 datasets (ablation) | EDT · D4RL sum | normalized return: 507.6 -> 570.3 | Improved |
| fixed history length 5 vs variable, summed over 8 datasets (ablation) | EDT · D4RL sum | normalized return: 483.9 -> 570.3 | Improved |
| lower expectile level alpha toward 0.5 (MSE) (ablation) | EDT · Hopper medium-replay | normalized return | Hurt |
| larger step size delta shrinking history search space (ablation) | EDT · D4RL locomotion | normalized return | Hurt |
| EDT vs DT in multi-task medium-replay locomotion (sum) | EDT-1 vs DT-1 · Multi-task D4RL | normalized return: 191.3 -> 276.4 | Improved |
| EDT vs IQL in multi-task medium-replay locomotion (sum) | EDT-1 vs IQL-1 · Multi-task D4RL | normalized return: 235.1 -> 276.4 | Improved |
| EDT vs DT on single-task hopper-medium-replay | EDT · D4RL hopper-medium-replay | normalized return: 61.9 -> 89.0 | Improved |
Closest papers by meaning
- Trust Guided Decision Transformer (NeurIPS 2026 · similarity 0.86)
- Rethinking Decision Transformer via Hierarchical Reinforcement Learning (ICML 2024 · similarity 0.86)
- ACT: Empowering Decision Transformer with Dynamic Programming via Advantage Conditioning (AAAI 2024 · similarity 0.85)
- Q-value Regularized Transformer for Offline Reinforcement Learning (ICML 2024 · similarity 0.85)
- In-Context Decision Transformer: Reinforcement Learning via Hierarchical Chain-of-Thought (ICML 2024 · similarity 0.84)
- Reinforcement Learning Gradients as Vitamin for Online Finetuning Decision Transformers (NeurIPS 2024 · similarity 0.84)
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.
