Enhancing Preference-based Linear Bandits via Human Response Time

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Li, Shen, Zhang, Yuyang, Ren, Zhaolin, Liang, Claire, Li, Na, Shah, Julie A.
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866909447029260288
author Li, Shen
Zhang, Yuyang
Ren, Zhaolin
Liang, Claire
Li, Na
Shah, Julie A.
author_facet Li, Shen
Zhang, Yuyang
Ren, Zhaolin
Liang, Claire
Li, Na
Shah, Julie A.
contents Interactive preference learning systems infer human preferences by presenting queries as pairs of options and collecting binary choices. Although binary choices are simple and widely used, they provide limited information about preference strength. To address this, we leverage human response times, which are inversely related to preference strength, as an additional signal. We propose a computationally efficient method that combines choices and response times to estimate human utility functions, grounded in the EZ diffusion model from psychology. Theoretical and empirical analyses show that for queries with strong preferences, response times complement choices by providing extra information about preference strength, leading to significantly improved utility estimation. We incorporate this estimator into preference-based linear bandits for fixed-budget best-arm identification. Simulations on three real-world datasets demonstrate that using response times significantly accelerates preference learning compared to choice-only approaches. Additional materials, such as code, slides, and talk video, are available at https://shenlirobot.github.io/pages/NeurIPS24.html
format Preprint
id arxiv_https___arxiv_org_abs_2409_05798
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Enhancing Preference-based Linear Bandits via Human Response Time
Li, Shen
Zhang, Yuyang
Ren, Zhaolin
Liang, Claire
Li, Na
Shah, Julie A.
Machine Learning
Artificial Intelligence
Human-Computer Interaction
Econometrics
Interactive preference learning systems infer human preferences by presenting queries as pairs of options and collecting binary choices. Although binary choices are simple and widely used, they provide limited information about preference strength. To address this, we leverage human response times, which are inversely related to preference strength, as an additional signal. We propose a computationally efficient method that combines choices and response times to estimate human utility functions, grounded in the EZ diffusion model from psychology. Theoretical and empirical analyses show that for queries with strong preferences, response times complement choices by providing extra information about preference strength, leading to significantly improved utility estimation. We incorporate this estimator into preference-based linear bandits for fixed-budget best-arm identification. Simulations on three real-world datasets demonstrate that using response times significantly accelerates preference learning compared to choice-only approaches. Additional materials, such as code, slides, and talk video, are available at https://shenlirobot.github.io/pages/NeurIPS24.html
title Enhancing Preference-based Linear Bandits via Human Response Time
topic Machine Learning
Artificial Intelligence
Human-Computer Interaction
Econometrics
url https://arxiv.org/abs/2409.05798