On Robust Cross Domain Alignment

TL;DR

This paper introduces three novel techniques to enhance the robustness of Gromov-Wasserstein distance for cross-domain alignment, addressing contamination issues and improving reliability in practical machine learning applications.

Contribution

It proposes specific robustification methods for GW that are tailored to its unique properties, unlike previous OT-based approaches.

Findings

Robust methods outperform existing techniques under contaminated data.

The proposed techniques maintain metric properties and provide robustness guarantees.

Empirical results show improved resilience in real ML tasks.

Abstract

The Gromov-Wasserstein (GW) distance is an effective measure of alignment between distributions supported on distinct ambient spaces. Calculating essentially the mutual departure from isometry, it has found vast usage in domain translation and network analysis. It has long been shown to be vulnerable to contamination in the underlying measures. All efforts to introduce robustness in GW have been inspired by similar techniques in optimal transport (OT), which predominantly advocate partial mass transport or unbalancing. In contrast, the cross-domain alignment problem being fundamentally different from OT, demands specific solutions to tackle diverse applications and contamination regimes. Deriving from robust statistics, we discuss three contextually novel techniques to robustify GW and its variants. For each method, we explore metric properties and robustness guarantees along with their co-dependencies and individual relations with the GW distance. For a comprehensive view, we empirically validate their superior resilience to contamination under real machine learning tasks against state-of-the-art methods.