Metric properties of partial and robust Gromov-Wasserstein distances

TL;DR

This paper investigates the theoretical properties of a relaxed Gromov-Wasserstein distance, introduces a new family of robust metrics inspired by classical distances, and demonstrates their robustness and topological equivalence to GW.

Contribution

It characterizes the metric failures of the relaxed GW, proposes a new family of robust partial GW distances, and proves their metric properties and robustness.

Findings

The relaxed GW fails to satisfy non-degeneracy and triangle inequality.

The new distances form true metrics and induce the same topology as GW.

The new metrics are more robust to outliers and perturbations.

Abstract

The Gromov-Wasserstein (GW) distances define a family of metrics, based on ideas from optimal transport, which enable comparisons between probability measures defined on distinct metric spaces. They are particularly useful in areas such as network analysis and geometry processing, as computation of a GW distance involves solving for registration between the objects which minimizes geometric distortion. Although GW distances have proven useful for various applications in the recent machine learning literature, it has been observed that they are inherently sensitive to outlier noise and cannot accommodate partial matching. This has been addressed by various constructions building on the GW framework; in this article, we focus specifically on a natural relaxation of the GW optimization problem, introduced by Chapel et al., which is aimed at addressing exactly these shortcomings. Our goal is to understand the theoretical properties of this relaxed optimization problem, from the viewpoint of metric geometry. While the relaxed problem fails to induce a metric, we derive precise characterizations of how it fails the axioms of non-degeneracy and triangle inequality. These observations lead us to define a novel family of distances, whose construction is inspired by the Prokhorov and Ky Fan distances, as well as by the recent work of Raghvendra et al.\ on robust versions of classical Wasserstein distance. We show that our new distances define true metrics, that they induce the same topology as the GW distances, and that they enjoy additional robustness to perturbations. These results provide a mathematically rigorous basis for using our robust partial GW distances in applications where outliers and partial matching are concerns.