AI Paper Atlas

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

Local Reinforcement Learning with Action-Conditioned Root Mean Squared Q-Functions

Zequan Wu, Mengye Ren

ICLR 2026 · Reinforcement learning › Policy gradient and actor-critic · approach: Temporal-difference and discounting theory · 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
add action conditioning at the input (ablation)ARQ · MinAtar Breakout and SpaceInvadersaverage return: ~85 vs. ~55Improved
add action conditioning at the input (ablation)AD · MinAtar Breakout and SpaceInvadersaverage return: slight improvementImproved
use RMS goodness instead of mean (ablation)ARQ · MinAtar Breakoutaverage return: 79.84 -> 87.84Improved
use RMS goodness instead of mean squared (ablation)ARQ · MinAtar Breakoutaverage return: 82.10 -> 87.84Improved
use RMS goodness instead of variance (ablation)ARQ · MinAtar SpaceInvadersaverage return: 416.46 -> 544.99Improved
use RMS goodness instead of mean squared (ablation)ARQ · MinAtar SpaceInvadersaverage return: 434.88 -> 544.99Improved
use RMS goodness instead of mean (ablation)ARQ · MinAtar SpaceInvadersaverage return: 500.13 -> 544.99Improved
replace AD scalar output with any ARQ goodness function (ablation)AD · MinAtar Breakoutaverage return: 67.40 -> 81.34Improved
replace AD with ARQ at 0.5x parameter scale (ablation)AD · MinAtar Breakoutaverage return: 66.34 -> 68.12Improved
replace AD with ARQ at 1x parameter scale (ablation)AD · MinAtar Breakoutaverage return: 64.20 -> 86.26Improved
replace AD with ARQ at 1.5x parameter scale (ablation)AD · MinAtar Breakoutaverage return: 56.63 -> 70.40Improved
replace AD with ARQ at 2x parameter scale (ablation)AD · MinAtar Breakoutaverage return: 59.79 -> 83.26Improved

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