Research map / Reinforcement learning / Offline RL
Algorithmic Guarantees for Distilling Supervised and Offline RL Datasets
Aaryan Gupta, Rishi Saket, Aravindan Raghuveer
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 |
|---|---|---|---|
| distill regression data by loss matching on sampled linear regressors instead of random subsampling | linear regression on Wine Quality and Boston Housing · red/white wine quality, Boston Housing | test MSE | Improved |
| use loss-matching distilled data instead of leverage-score subsampling | linear regression · red/white wine quality, Boston Housing | test MSE | Mixed |
| train on the distilled set instead of the full training set | linear regression · Wine Quality, Boston Housing | test MSE | No clear effect |
| distill offline RL data by Bellman-loss matching instead of random subsampling | Fitted-Q Iteration with 2-layer networks · Cartpole, Mountain Car, Acrobot | max evaluation return | Improved |
| increase synthetic dataset size N_syn (ablation) | Bellman-loss distillation · Cartpole, Acrobot | max evaluation return | Improved |
Closest papers by meaning
- Towards Lossless Dataset Distillation via Difficulty-Aligned Trajectory Matching (ICLR 2024 · similarity 0.89)
- Embarrassingly Simple Dataset Distillation (ICLR 2024 · similarity 0.89)
- Self-Supervised Dataset Distillation for Transfer Learning (ICLR 2024 · similarity 0.88)
- Group Distributionally Robust Dataset Distillation with Risk Minimization (ICLR 2025 · similarity 0.88)
- Efficient Dataset Distillation for Pre-Trained Self-Supervised Models via Statistical Flow Matching (NeurIPS 2026 · similarity 0.88)
- Dataset Distillation for Pre-Trained Self-Supervised Vision Models (NeurIPS 2025 · similarity 0.87)
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.
