Enregistré dans:
Détails bibliographiques
Auteurs principaux: Liu, Song, Yu, Jiahao, Simons, Jack, Yi, Mingxuan, Beaumont, Mark
Format: Preprint
Publié: 2023
Sujets:
Accès en ligne:https://arxiv.org/abs/2305.15577
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866909217779089408
author Liu, Song
Yu, Jiahao
Simons, Jack
Yi, Mingxuan
Beaumont, Mark
author_facet Liu, Song
Yu, Jiahao
Simons, Jack
Yi, Mingxuan
Beaumont, Mark
contents Many machine learning problems can be seen as approximating a \textit{target} distribution using a \textit{particle} distribution by minimizing their statistical discrepancy. Wasserstein Gradient Flow can move particles along a path that minimizes the $f$-divergence between the target and particle distributions. To move particles, we need to calculate the corresponding velocity fields derived from a density ratio function between these two distributions. Previous works estimated such density ratio functions and then differentiated the estimated ratios. These approaches may suffer from overfitting, leading to a less accurate estimate of the velocity fields. Inspired by non-parametric curve fitting, we directly estimate these velocity fields using interpolation techniques. We prove that our estimators are consistent under mild conditions. We validate their effectiveness using novel applications on domain adaptation and missing data imputation.
format Preprint
id arxiv_https___arxiv_org_abs_2305_15577
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Minimizing $f$-Divergences by Interpolating Velocity Fields
Liu, Song
Yu, Jiahao
Simons, Jack
Yi, Mingxuan
Beaumont, Mark
Machine Learning
Many machine learning problems can be seen as approximating a \textit{target} distribution using a \textit{particle} distribution by minimizing their statistical discrepancy. Wasserstein Gradient Flow can move particles along a path that minimizes the $f$-divergence between the target and particle distributions. To move particles, we need to calculate the corresponding velocity fields derived from a density ratio function between these two distributions. Previous works estimated such density ratio functions and then differentiated the estimated ratios. These approaches may suffer from overfitting, leading to a less accurate estimate of the velocity fields. Inspired by non-parametric curve fitting, we directly estimate these velocity fields using interpolation techniques. We prove that our estimators are consistent under mild conditions. We validate their effectiveness using novel applications on domain adaptation and missing data imputation.
title Minimizing $f$-Divergences by Interpolating Velocity Fields
topic Machine Learning
url https://arxiv.org/abs/2305.15577