AI Paper Atlas

Research map / Reinforcement learning / Policy gradient and actor-critic

Zero-Shot Off-Policy Learning

Arip Asadulaev, Maksim Bobrin, Salem Lahlou, Dmitry V. Dylov, Fakhri Karray, Martin Takáč

ICML 2026 · Reinforcement learning › Policy gradient and actor-critic · approach: Model-based RL with guarantees · 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
replace FB latent inference with ZOL occupancy-ratio latent optimizationFB · ExORL DMC-state avgreturn: 592.5 -> 627.2Improved
replace FB latent inference with ZOLFB · OGBench avgsuccess rate: 0.45 -> 0.47Improved
ZOL vs FB on puzzle-3x3-playFB · OGBench puzzle-3x3-playsuccess rate: 0.0 -> 0.14Improved
ZOL vs ReLA on antmaze-large-navigateReLA · OGBench antmaze-large-navigatesuccess rate: 0.77 -> 0.63Hurt
ZOL vs FB on antmaze-medium-navigateFB · OGBench antmaze-medium-navigatesuccess rate: 0.95 -> 0.91Hurt
ZOL vs ReLA on Quadruped standReLA · ExORL Quadruped standreturn: 846 -> 660Hurt
ZOL vs TD-JEPA on Quadruped walkTD-JEPA · ExORL Quadruped walkreturn: 600 -> 411Hurt
ZOL vs FB on Point Mass bottom rightFB · ExORL Point Mass maze bottom rightreturn: 165 -> 225Improved
use fewer inner optimization steps (ablation)ZOL · DMC (normalized return)normalized returnImproved
add small trust-region penalty (ablation)ZOL · DMC (normalized return)normalized returnImproved
increase chi2 penalty to large value (ablation)ZOL · DMC (normalized return)normalized returnHurt
vary weight clipping value (ablation)ZOL · DMC (normalized return)normalized returnNo 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