Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Li, Yuxuan, Kethireddy, Harshith Reddy, Das, Srijita
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2601.01904
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866911524020289536
author Li, Yuxuan
Kethireddy, Harshith Reddy
Das, Srijita
author_facet Li, Yuxuan
Kethireddy, Harshith Reddy
Das, Srijita
contents Learning from Preferences in Reinforcement Learning (PbRL) has gained attention recently, as it serves as a natural fit for complicated tasks where the reward function is not easily available. However, preferences often come with uncertainty and noise if they are not from perfect teachers. Much prior literature aimed to detect noise, but with limited types of noise and most being uniformly distributed with no connection to observations. In this work, we formalize the notion of targeted feature-dependent noise and propose several variants like trajectory feature noise, trajectory similarity noise, margin dependent noise, and Language Model noise. We evaluate feature-dependent noise, where noise is correlated with certain features in complex continuous control tasks from DMControl and Meta-world. Our experiments show that in some feature-dependent noise settings, the state-of-the-art noise-robust PbRL method's learning performance is significantly deteriorated, while PbRL method with no explicit denoising can surprisingly outperform noise-robust PbRL in the majority of settings. We also find language models' noise exhibits similar characteristics to feature-dependent noise, thereby simulating realistic humans and call for further study in learning with feature-dependent noise robustly.
format Preprint
id arxiv_https___arxiv_org_abs_2601_01904
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Evaluating Feature Dependent Noise in Preference-based Reinforcement Learning
Li, Yuxuan
Kethireddy, Harshith Reddy
Das, Srijita
Machine Learning
Artificial Intelligence
Learning from Preferences in Reinforcement Learning (PbRL) has gained attention recently, as it serves as a natural fit for complicated tasks where the reward function is not easily available. However, preferences often come with uncertainty and noise if they are not from perfect teachers. Much prior literature aimed to detect noise, but with limited types of noise and most being uniformly distributed with no connection to observations. In this work, we formalize the notion of targeted feature-dependent noise and propose several variants like trajectory feature noise, trajectory similarity noise, margin dependent noise, and Language Model noise. We evaluate feature-dependent noise, where noise is correlated with certain features in complex continuous control tasks from DMControl and Meta-world. Our experiments show that in some feature-dependent noise settings, the state-of-the-art noise-robust PbRL method's learning performance is significantly deteriorated, while PbRL method with no explicit denoising can surprisingly outperform noise-robust PbRL in the majority of settings. We also find language models' noise exhibits similar characteristics to feature-dependent noise, thereby simulating realistic humans and call for further study in learning with feature-dependent noise robustly.
title Evaluating Feature Dependent Noise in Preference-based Reinforcement Learning
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2601.01904