AI Paper Atlas

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

Entropy-preserving reinforcement learning

Aleksei Petrenko, Ben Lipkin, Kevin Chen, Erik Wijmans, Marco Francis Cusumano-Towner, Raja Giryes, Philipp Kraehenbuehl

ICLR 2026 · Reinforcement learning › Policy gradient and actor-critic · approach: Alternative policy update rules · 9 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 adaptive entropy control (REPO-R) on top of GRPOGRPO · AppWorld and AIMEtest accuracyImproved
add adaptive asymmetric clipping (ADAPO) on top of DAPODAPO · AppWorld and AIMEtest accuracyImproved
use FP16 instead of BF16 training with log-prob rounding fix (ablation)DAPO · AppWorld (Qwen-3-8B)entropy dynamicsImproved
switch to FP16 training for strictly on-policy RLOO (ablation)RLOO · AppWorld (Qwen-3-32B)task goal completion: 79% Test Normal, 71% Test ChallengeImproved
use PPO-style off-policy multi-epoch updates (GRPO) instead of strictly on-policy RLOORLOO · AppWorld/AIMEentropy retention: GRPO reduces entropy by nearly 90%Hurt
use asymmetric clipping (DAPO) and sequence-level clipping (GSPO)GRPO/LOOP · AppWorld/AIMEentropy retentionImproved
train on entropy-preserving algorithms before sequential training on a new environmentGRPO · Sequential AIME to AppWorld and backtest accuracy (TGC)Improved

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