Entropic Optimal Transport in Random Graphs

Abstract

In graph analysis, a classic task consists in computing similarity measures between (groups of) nodes. In latent space random graphs, nodes are associated to unknown latent variables. One may then seek to compute distances directly in the latent space, using only the graph structure. In this paper, we show that it is possible to consistently estimate entropic-regularized Optimal Transport (OT) distances between groups of nodes in the latent space. We provide a general stability result for entropic OT with respect to perturbations of the cost matrix. We then apply it to several examples of random graphs, such as graphons or $\epsilon$-graphs on manifolds. Along the way, we prove new concentration results for the so-called Universal Singular Value Thresholding estimator, and for the estimation of geodesic distances on a manifold.

References in corpus (7)

• Consistency of spectral clustering
• Matrix estimation by Universal Singular Value Thresholding
• Stochastic Optimization for Large-scale Optimal Transport
• Continuum limit of total variation on point clouds
• Convergence and Stability of Graph Convolutional Networks on Large Random Graphs
• Latent Distance Estimation for Random Geometric Graphs
• Sparse and Smooth: improved guarantees for Spectral Clustering in the Dynamic Stochastic Block Model