Saved in:
Bibliographic Details
Main Authors: Tarasov, Roman, Mokrov, Petr, Gazdieva, Milena, Burnaev, Evgeny, Korotin, Alexander
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2502.01310
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908849086136320
author Tarasov, Roman
Mokrov, Petr
Gazdieva, Milena
Burnaev, Evgeny
Korotin, Alexander
author_facet Tarasov, Roman
Mokrov, Petr
Gazdieva, Milena
Burnaev, Evgeny
Korotin, Alexander
contents Neural network-based optimal transport (OT) is a recent and fruitful direction in the generative modeling community. It finds its applications in various fields such as domain translation, image super-resolution, computational biology and others. Among the existing OT approaches, of considerable interest are adversarial minimax solvers based on semi-dual formulations of OT problems. While promising, these methods lack theoretical investigation from a statistical learning perspective. Our work fills this gap by establishing upper bounds on the generalization error of an approximate OT map recovered by the minimax quadratic OT solver. Importantly, the bounds we derive depend solely on some standard statistical and mathematical properties of the considered functional classes (neural nets). While our analysis focuses on the quadratic OT, we believe that similar bounds could be derived for general OT case, paving the promising direction for future research. Our experimental illustrations are available online https://github.com/milenagazdieva/StatOT.
format Preprint
id arxiv_https___arxiv_org_abs_2502_01310
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A Statistical Learning Perspective on Semi-dual Adversarial Neural Optimal Transport Solvers
Tarasov, Roman
Mokrov, Petr
Gazdieva, Milena
Burnaev, Evgeny
Korotin, Alexander
Machine Learning
Artificial Intelligence
Neural network-based optimal transport (OT) is a recent and fruitful direction in the generative modeling community. It finds its applications in various fields such as domain translation, image super-resolution, computational biology and others. Among the existing OT approaches, of considerable interest are adversarial minimax solvers based on semi-dual formulations of OT problems. While promising, these methods lack theoretical investigation from a statistical learning perspective. Our work fills this gap by establishing upper bounds on the generalization error of an approximate OT map recovered by the minimax quadratic OT solver. Importantly, the bounds we derive depend solely on some standard statistical and mathematical properties of the considered functional classes (neural nets). While our analysis focuses on the quadratic OT, we believe that similar bounds could be derived for general OT case, paving the promising direction for future research. Our experimental illustrations are available online https://github.com/milenagazdieva/StatOT.
title A Statistical Learning Perspective on Semi-dual Adversarial Neural Optimal Transport Solvers
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2502.01310