Bandit Pareto Set Identification in a Multi-Output Linear Model

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kone, Cyrille, Kaufmann, Emilie, Richert, Laura
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915373312376832
author Kone, Cyrille
Kaufmann, Emilie
Richert, Laura
author_facet Kone, Cyrille
Kaufmann, Emilie
Richert, Laura
contents We study the Pareto Set Identification (PSI) problem in a structured multi-output linear bandit model. In this setting, each arm is associated a feature vector belonging to $\mathbb{R}^h$, and its mean vector in $\mathbb{R}^d$ linearly depends on this feature vector through a common unknown matrix $Θ\in \mathbb{R}^{h \times d}$. The goal is to identify the set of non-dominated arms by adaptively collecting samples from the arms. We introduce and analyze the first optimal design-based algorithms for PSI, providing nearly optimal guarantees in both the fixed-budget and the fixed-confidence settings. Notably, we show that the difficulty of these tasks mainly depends on the sub-optimality gaps of $h$ arms only. Our theoretical results are supported by an extensive benchmark on synthetic and real-world datasets.
format Preprint
id arxiv_https___arxiv_org_abs_2507_04255
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Bandit Pareto Set Identification in a Multi-Output Linear Model
Kone, Cyrille
Kaufmann, Emilie
Richert, Laura
Machine Learning
We study the Pareto Set Identification (PSI) problem in a structured multi-output linear bandit model. In this setting, each arm is associated a feature vector belonging to $\mathbb{R}^h$, and its mean vector in $\mathbb{R}^d$ linearly depends on this feature vector through a common unknown matrix $Θ\in \mathbb{R}^{h \times d}$. The goal is to identify the set of non-dominated arms by adaptively collecting samples from the arms. We introduce and analyze the first optimal design-based algorithms for PSI, providing nearly optimal guarantees in both the fixed-budget and the fixed-confidence settings. Notably, we show that the difficulty of these tasks mainly depends on the sub-optimality gaps of $h$ arms only. Our theoretical results are supported by an extensive benchmark on synthetic and real-world datasets.
title Bandit Pareto Set Identification in a Multi-Output Linear Model
topic Machine Learning
url https://arxiv.org/abs/2507.04255