Saved in:
Bibliographic Details
Main Authors: Sotiropoulou, Athina, Alvarez-Melis, David
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2407.14957
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910536740896768
author Sotiropoulou, Athina
Alvarez-Melis, David
author_facet Sotiropoulou, Athina
Alvarez-Melis, David
contents Optimal Transport (OT) has recently emerged as a powerful framework for learning minimal-displacement maps between distributions. The predominant approach involves a neural parametrization of the Monge formulation of OT, typically assuming the same space for both distributions. However, the setting across ``incomparable spaces'' (e.g., of different dimensionality), corresponding to the Gromov- Wasserstein distance, remains underexplored, with existing methods often imposing restrictive assumptions on the cost function. In this paper, we present a novel neural formulation of the Gromov-Monge (GM) problem rooted in one of its fundamental properties: invariance to strong isomorphisms. We operationalize this property by decomposing the learnable OT map into two components: (i) an approximate strong isomorphism between the source distribution and an intermediate reference distribution, and (ii) a GM-optimal map between this reference and the target distribution. Our formulation leverages and extends the Monge gap regularizer of Uscidda & Cuturi (2023) to eliminate the need for complex architectural requirements of other neural OT methods, yielding a simple but practical method that enjoys favorable theoretical guarantees. Our preliminary empirical results show that our framework provides a promising approach to learn OT maps across diverse spaces.
format Preprint
id arxiv_https___arxiv_org_abs_2407_14957
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Strongly Isomorphic Neural Optimal Transport Across Incomparable Spaces
Sotiropoulou, Athina
Alvarez-Melis, David
Machine Learning
Optimal Transport (OT) has recently emerged as a powerful framework for learning minimal-displacement maps between distributions. The predominant approach involves a neural parametrization of the Monge formulation of OT, typically assuming the same space for both distributions. However, the setting across ``incomparable spaces'' (e.g., of different dimensionality), corresponding to the Gromov- Wasserstein distance, remains underexplored, with existing methods often imposing restrictive assumptions on the cost function. In this paper, we present a novel neural formulation of the Gromov-Monge (GM) problem rooted in one of its fundamental properties: invariance to strong isomorphisms. We operationalize this property by decomposing the learnable OT map into two components: (i) an approximate strong isomorphism between the source distribution and an intermediate reference distribution, and (ii) a GM-optimal map between this reference and the target distribution. Our formulation leverages and extends the Monge gap regularizer of Uscidda & Cuturi (2023) to eliminate the need for complex architectural requirements of other neural OT methods, yielding a simple but practical method that enjoys favorable theoretical guarantees. Our preliminary empirical results show that our framework provides a promising approach to learn OT maps across diverse spaces.
title Strongly Isomorphic Neural Optimal Transport Across Incomparable Spaces
topic Machine Learning
url https://arxiv.org/abs/2407.14957