AI Paper Atlas

Research map / Reinforcement learning / Offline RL

Variational OOD State Correction for Offline Reinforcement Learning

Ke Jiang, Wen Jiang, Xiaoyang Tan

AAAI 2026 · Reinforcement learning › Offline RL · approach: Offline RL sample complexity theory · 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
add DASP density-aware regularizer to offline actoroffline RL agent · D4RL MuJoCo averagenormalized score: 86.2 -> 89.0 (vs OSR-10)Improved
use DASP on AntMaze instead of best baselineoffline RL agent · D4RL AntMaze averagenormalized score: 68.5 -> 68.6 (vs SCAS)No clear effect
DASP vs OSR-10 under large external forceDASP · OOSMuJoCo largenormalized return: Halfcheetah 50.8 -> 54.6; Hopper 94.7 -> 89.5Mixed
increase DASP weight alpha from 0.01 to 0.1 (ablation)DASP · Halfcheetah-mediumnormalized score: 66.0 -> 70.4Improved
raise alpha too high (0.1 to 100) (ablation)DASP · Halfcheetah-mediumnormalized score: 70.4 -> 51.6Hurt
raise alpha too high (0.05 to 100) (ablation)DASP · Walker2d-mediumnormalized score: 108.6 -> 85.6Hurt
alpha 0.1 vs 0.5 on hopper (ablation)DASP · Hopper-mediumnormalized score: 108.6 -> 105.1No clear effect
use learned DASP model instead of true dynamics model for safety scoreDASP term · Halfcheetah/Hopper/Walker2dsafety score gap safe vs unsafe: TDM 0.61 vs 0.37; DASP 0.64 vs 0.38 (Halfcheetah)No clear effect

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.

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

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