Adaptive learning of density ratios in RKHS

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zellinger, Werner, Kindermann, Stefan, Pereverzyev, Sergei V.
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916153559875584
author Zellinger, Werner
Kindermann, Stefan
Pereverzyev, Sergei V.
author_facet Zellinger, Werner
Kindermann, Stefan
Pereverzyev, Sergei V.
contents Estimating the ratio of two probability densities from finitely many observations of the densities is a central problem in machine learning and statistics with applications in two-sample testing, divergence estimation, generative modeling, covariate shift adaptation, conditional density estimation, and novelty detection. In this work, we analyze a large class of density ratio estimation methods that minimize a regularized Bregman divergence between the true density ratio and a model in a reproducing kernel Hilbert space (RKHS). We derive new finite-sample error bounds, and we propose a Lepskii type parameter choice principle that minimizes the bounds without knowledge of the regularity of the density ratio. In the special case of quadratic loss, our method adaptively achieves a minimax optimal error rate. A numerical illustration is provided.
format Preprint
id arxiv_https___arxiv_org_abs_2307_16164
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Adaptive learning of density ratios in RKHS
Zellinger, Werner
Kindermann, Stefan
Pereverzyev, Sergei V.
Machine Learning
Statistics Theory
68T05, 68Q32
Estimating the ratio of two probability densities from finitely many observations of the densities is a central problem in machine learning and statistics with applications in two-sample testing, divergence estimation, generative modeling, covariate shift adaptation, conditional density estimation, and novelty detection. In this work, we analyze a large class of density ratio estimation methods that minimize a regularized Bregman divergence between the true density ratio and a model in a reproducing kernel Hilbert space (RKHS). We derive new finite-sample error bounds, and we propose a Lepskii type parameter choice principle that minimizes the bounds without knowledge of the regularity of the density ratio. In the special case of quadratic loss, our method adaptively achieves a minimax optimal error rate. A numerical illustration is provided.
title Adaptive learning of density ratios in RKHS
topic Machine Learning
Statistics Theory
68T05, 68Q32
url https://arxiv.org/abs/2307.16164