The Unbalanced Gromov Wasserstein Distance: Conic Formulation and Relaxation

09/09/2020
by   Thibault Séjourné, et al.
5

Comparing metric measure spaces (i.e. a metric space endowed with a probability distribution) is at the heart of many machine learning problems. This includes for instance predicting properties of molecules in quantum chemistry or generating graphs with varying connectivity. The most popular distance between such metric measure spaces is the Gromov-Wasserstein (GW) distance, which is the solution of a quadratic assignment problem. This distance has been successfully applied to supervised learning and generative modeling, for applications as diverse as quantum chemistry or natural language processing. The GW distance is however limited to the comparison of metric measure spaces endowed with a probability distribution. This strong limitation is problematic for many applications in ML where there is no a priori natural normalization on the total mass of the data. Furthermore, imposing an exact conservation of mass across spaces is not robust to outliers and often leads to irregular matching. To alleviate these issues, we introduce two Unbalanced Gromov-Wasserstein formulations: a distance and a more computationally tractable upper-bounding relaxation. They both allow the comparison of metric spaces equipped with arbitrary positive measures up to isometries.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
02/24/2018

Minimax Distribution Estimation in Wasserstein Distance

The Wasserstein metric is an important measure of distance between proba...
research
02/20/2018

Free complete Wasserstein algebras

We present an algebraic account of the Wasserstein distances W_p on comp...
research
07/25/2023

Computing the Gromov–Hausdorff distance using first-order methods

The Gromov–Hausdorff distance measures the difference in shape between c...
research
06/04/2022

Concentration of the missing mass in metric spaces

We study the estimation of the probability to observe data further than ...
research
04/24/2022

Design and analysis of computer experiments with both numeral and distribution inputs

Nowadays stochastic computer simulations with both numeral and distribut...
research
09/25/2019

Optimal Transport to a Variety

We study the problem of minimizing the Wasserstein distance between a pr...
research
02/05/2020

Fast and Robust Comparison of Probability Measures in Heterogeneous Spaces

The problem of comparing distributions endowed with their own geometry a...

Please sign up or login with your details

Forgot password? Click here to reset