Saved in:
Bibliographic Details
Main Authors: Weissmann, Simon, Freihaut, Till, Vernade, Claire, Ramponi, Giorgia, Döring, Leif
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2503.02735
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912258481717248
author Weissmann, Simon
Freihaut, Till
Vernade, Claire
Ramponi, Giorgia
Döring, Leif
author_facet Weissmann, Simon
Freihaut, Till
Vernade, Claire
Ramponi, Giorgia
Döring, Leif
contents In the context of stochastic bandit models, this article examines how to design sample-efficient behavior policies for the importance sampling evaluation of multiple target policies. From importance sampling theory, it is well established that sample efficiency is highly sensitive to the KL divergence between the target and importance sampling distributions. We first analyze a single behavior policy defined as the KL-barycenter of the target policies. Then, we refine this approach by clustering the target policies into groups with small KL divergences and assigning each cluster its own KL-barycenter as a behavior policy. This clustered KL-based policy evaluation (CKL-PE) algorithm provides a novel perspective on optimal policy selection. We prove upper bounds on the sample complexity of our method and demonstrate its effectiveness with numerical validation.
format Preprint
id arxiv_https___arxiv_org_abs_2503_02735
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Clustered KL-barycenter design for policy evaluation
Weissmann, Simon
Freihaut, Till
Vernade, Claire
Ramponi, Giorgia
Döring, Leif
Machine Learning
In the context of stochastic bandit models, this article examines how to design sample-efficient behavior policies for the importance sampling evaluation of multiple target policies. From importance sampling theory, it is well established that sample efficiency is highly sensitive to the KL divergence between the target and importance sampling distributions. We first analyze a single behavior policy defined as the KL-barycenter of the target policies. Then, we refine this approach by clustering the target policies into groups with small KL divergences and assigning each cluster its own KL-barycenter as a behavior policy. This clustered KL-based policy evaluation (CKL-PE) algorithm provides a novel perspective on optimal policy selection. We prove upper bounds on the sample complexity of our method and demonstrate its effectiveness with numerical validation.
title Clustered KL-barycenter design for policy evaluation
topic Machine Learning
url https://arxiv.org/abs/2503.02735