AI Paper Atlas

Research map / Reinforcement learning / Offline RL

Algorithmic Guarantees for Distilling Supervised and Offline RL Datasets

Aaryan Gupta, Rishi Saket, Aravindan Raghuveer

ICLR 2026 · Reinforcement learning › Offline RL · approach: Simple practical offline RL · 0 citations (OpenAlex)

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 changedWhereMeasuredEffect
distill regression data by loss matching on sampled linear regressors instead of random subsamplinglinear regression on Wine Quality and Boston Housing · red/white wine quality, Boston Housingtest MSEImproved
use loss-matching distilled data instead of leverage-score subsamplinglinear regression · red/white wine quality, Boston Housingtest MSEMixed
train on the distilled set instead of the full training setlinear regression · Wine Quality, Boston Housingtest MSENo clear effect
distill offline RL data by Bellman-loss matching instead of random subsamplingFitted-Q Iteration with 2-layer networks · Cartpole, Mountain Car, Acrobotmax evaluation returnImproved
increase synthetic dataset size N_syn (ablation)Bellman-loss distillation · Cartpole, Acrobotmax evaluation returnImproved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy