AI Paper Atlas

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

Trust Region-Guided Proximal Policy Optimization

Yuhui Wang, Hao He, Xiaoyang Tan, Yaozhong Gan

NeurIPS 2019 · Reinforcement learning › Policy gradient and actor-critic · approach: Tree search and policy optimization · 30 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 constant PPO clipping with trust-region-guided adaptive clippingPPO · Humanoidtimesteps to threshold (5000): 7241 -> 4653Improved
Replace constant PPO clipping with trust-region-guided adaptive clippingPPO · Reacheraverage reward, last 40% episodes: -6.7 -> -7.9Hurt
Replace constant PPO clipping with trust-region-guided adaptive clippingPPO · Swimmertimesteps to threshold (90): 564 -> 353.0Improved
Replace constant PPO clipping with trust-region-guided adaptive clippingPPO · HalfCheetahaverage reward, last 40% episodes: 4600.2 -> 4986.1Improved
Replace constant PPO clipping with trust-region-guided adaptive clippingPPO · Hopperaverage reward, last 40% episodes: 2848.9 -> 3200.5Improved
Replace constant PPO clipping with trust-region-guided adaptive clippingPPO · Walker2daverage reward, last 40% episodes: 3276.2 -> 3886.8Improved
Use trust-region-guided adaptive clipping instead of PPO clippingPPO · continuous-armed bandittrials trapped in local optima: 20% -> almost all trials found the optimal solutionImproved
Use trust-region-guided adaptive clipping instead of PPO clippingPPO · discrete-armed bandittrials trapped in local optima: 30% -> almost all trials found the optimal solutionImproved
Increase PPO clipping parameter from epsilon=0.2 to epsilon=0.6PPO · MuJoCo benchmark taskstask performanceHurt
Add adaptive clipping rather than use constant PPO clippingPPO · MuJoCo benchmark taskspolicy entropyImproved

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