AI Paper Atlas

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

V-MPO: On-Policy Maximum a Posteriori Policy Optimization for Discrete and Continuous Control

H. Francis Song, Abbas Abdolmaleki, Jost Tobias Springenberg, Aidan Clark, Hubert Soyer, Jack W. Rae, Seb Noury, Arun Ahuja, Siqi Liu, Dhruva Tirumala, Nicolas Heess, Dan Belov and 2 more

ICLR 2020 · Reinforcement learning › Policy gradient and actor-critic · approach: Trajectory reuse in policy gradients · 51 citations (OpenAlex)

Read the paper

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
Use V-MPO instead of IMPALAAtari-57 multi-task · Atari-57human-normalized median: 117.56 -> 155.70Improved
Use the top 50% of batch advantages for the E-step (ablation)V-MPO · policy improvementlearning performanceImproved
Remove the parametric KL constraint (ablation)V-MPO on humanoid gaps from pixels · Humanoid gapslearning performanceHurt
Use V-MPO rather than previously reported methods22-dimensional humanoid from full-state observations · Humanoid runasymptotic returnImproved
Use V-MPO rather than previously reported methodsOpenAI Gym tasks · OpenAI Gymasymptotic performanceImproved

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