AI Paper Atlas

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

Bayesian Policy Gradient and Actor-Critic Algorithms

Mohammad Ghavamzadeh, Yaakov Engel, Michal Valko

JMLR 2016 · Reinforcement learning › Policy gradient and actor-critic · approach: Trajectory reuse in policy gradients

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
use Bayesian quadrature instead of Monte-Carlo gradient estimate (10 samples)policy gradient estimation · Bandit r(a)=astd of gradient estimate: 0.438 -> 0.050Improved
use Bayesian quadrature instead of Monte-Carlo gradient estimate (100 samples)policy gradient estimation · Bandit r(a)=astd of gradient estimate: 0.140 -> 0.000001Improved
use Bayesian quadrature instead of Monte-Carlo gradient estimate (10 samples)policy gradient estimation · Bandit r(a)=a^2std of gradient estimate: 2.831 -> 0.226Improved
use Bayesian quadrature instead of Monte-Carlo gradient estimate (100 samples)policy gradient estimation · Bandit r(a)=a^2std of gradient estimate: 0.857 -> 0.000011Improved
use BQ instead of MC estimatepolicy gradient estimation · LQR gradient estimationMSE of gradientImproved
add Gaussian reward noise (sigma_r = 0.1 or 1)BQ and MC gradient estimates · LQR gradient estimationgradient errorHurt
use BPG/BPNG instead of MCPG for sample size M = 5policy optimization · LQRaverage expected returnHurt
use BPG/BPNG instead of MCPG for sample sizes M = 10, 20, 40policy optimization · LQRaverage expected returnImproved

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