AI Paper Atlas

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

GIPO: Gaussian Importance Sampling Policy Optimization

Chengxuan Lu, Zhenquan Zhang, Shukuan Wang, Qunzhi Lin, Baigui Sun, Yang Liu

ICML 2026 · Reinforcement learning › Policy gradient and actor-critic · approach: Confounding-robust policy evaluation · 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 PPO hard clipping with Gaussian log-ratio trust weight (GIPO)PPO-style actor-critic with stale replay · Meta-World (10 tasks, stale regime)IQM normalized score: 0.180 -> 0.730Improved
use symmetric Gaussian weight instead of SAPO smooth weightingGIPO · Meta-World (10 tasks, stale regime)IQM normalized score: 0.412 -> 0.730Improved
use different sigma for positive and negative advantages (advantage-aware) (ablation)GIPO · Meta-World (10 tasks, stale regime)IQM normalized score: 0.730 (1.0,1.0) vs 0.543 (1.0,0.5)No clear effect
use advantage-aware sigma on MuJoCo (ablation)GIPO · MuJoCo (10 tasks)IQM normalized score: 0.606 -> 0.602No clear effect
advantage-aware sigma (0.2, 0.1) instead of (0.2, 0.2) (ablation)GIPO · MuJoCo Hopperaverage return: 786.9 -> 1184.1Improved
increase sigma from 0.5 to 1.0 (weaker damping) (ablation)GIPO · MuJoCo (10 tasks)IQM normalized score: 0.606 -> 0.478Hurt
increase sigma from 0.2 to 1.0 on Ant (ablation)GIPO · MuJoCo Antaverage return: 977.1 -> 20.4Hurt
best sigma depends on setting (1.0 on Meta-World stale, 0.5 on MuJoCo) (ablation)GIPO · Meta-World vs MuJoCoIQM normalized score: Meta-World best 0.730 at sigma 1.0; MuJoCo 0.478 at sigma 1.0Mixed
use GIPO instead of PPO on Classic ControlPPO · Classic Control (4 tasks)IQM normalized score: 0.7908 -> 0.8095Improved
use GIPO (0.2) instead of PPO on PendulumPPO · Pendulum-v1final rollout average return: -215.0 -> -276.8Hurt
use GIPO in fresh data regimePPO-Clip and SAPO · Meta-World (fresh regime)episodic returnNo clear effect
use GIPO with 3 actors per trainer (staler data)PPO and SAPO · LIBERO (OpenVLA-OFT 7B)average success rateImproved

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