AI Paper Atlas

Research map / Reinforcement learning / RL from feedback

Reward Learning through Ranking Mean Squared Error

Chaitanya Kharyal, Calarina Muslimani, Matthew E. Taylor

ICML 2026 · Reinforcement learning › RL from feedback · approach: Theory of RLHF · 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 the RbRL cross-entropy loss with the ranking MSE (rMSE) loss (ablation)SAC with learned reward · offline feedback: Reacher, Inverted Double Pendulum, Half CheetahSAC return / learning speed: faster learning or higher final return in all domains (p < 0.03)Improved
use R4 instead of RbRL and preference methods (PEBBLE, SURF, QPA) in online feedbackSAC on DMC · six DMC environmentsSAC return: faster learning in 3 of 6 environments, higher final return in 4 of 6 (p < 0.0125)Improved
use R4 instead of RbRL on real human ratingsR4 reward learning · Reacher (human ratings)final return: -44.62 -> -27.65Improved
use R4 instead of RbRL on real human ratingsR4 reward learning · Hopper (human ratings)final return: 723.87 -> 1410.10Improved
train R4 on human ratings instead of simulated ratingsR4 · ReacherSAC return: small gapHurt
remove dynamic feedback schedule and stratified sampling from R4 (ablation)R4 online · three online DMC domainsSAC returnNo clear effect
give baselines the dynamic feedback schedule (ablation)RbRL, PEBBLE, SURF, QPA · online DMCSAC returnHurt
inject label noise into ratings (reassign eta% of trajectories to true bin +/-1) (ablation)R4 vs RbRL · Inverted Double Pendulum (offline)SAC return: R4 at 80% noise comparable to RbRL at 10% noiseMixed
vary the number of rating bins (ablation)R4 vs RbRL · ReacherSAC returnNo clear effect
vary soft-rank regularization strength (ablation)R4 · Inverted Double Pendulumfinal return: 83% of runs with strengths 0.065 to 1 succeedNo clear effect

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