Saved in:
Bibliographic Details
Main Author: Nagpal, Chirag
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2508.13189
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916905940418560
author Nagpal, Chirag
author_facet Nagpal, Chirag
contents Approaches for estimating preferences from human annotated data typically involves inducing a distribution over a ranked list of choices such as the Plackett-Luce model. Indeed, modern AI alignment tools such as Reward Modelling and Direct Preference Optimization are based on the statistical assumptions posed by the Plackett-Luce model. In this paper, I will connect the Plackett-Luce model to another classical and well known statistical model, the Cox Proportional Hazards model and attempt to shed some light on the implications of the connection therein.
format Preprint
id arxiv_https___arxiv_org_abs_2508_13189
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Preference Models assume Proportional Hazards of Utilities
Nagpal, Chirag
Machine Learning
Artificial Intelligence
Approaches for estimating preferences from human annotated data typically involves inducing a distribution over a ranked list of choices such as the Plackett-Luce model. Indeed, modern AI alignment tools such as Reward Modelling and Direct Preference Optimization are based on the statistical assumptions posed by the Plackett-Luce model. In this paper, I will connect the Plackett-Luce model to another classical and well known statistical model, the Cox Proportional Hazards model and attempt to shed some light on the implications of the connection therein.
title Preference Models assume Proportional Hazards of Utilities
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2508.13189