Convergence of Manifold Filter-Combine Networks

TL;DR

It is proved that the method for implementing Manifold Filter-Combine Networks on high-dimensional point clouds is consistent in the sense that it converges to a continuum limit as the number of data points tends to infinity.

Abstract

In order to better understand manifold neural networks (MNNs), we introduce Manifold Filter-Combine Networks (MFCNs). The filter-combine framework parallels the popular aggregate-combine paradigm for graph neural networks (GNNs) and naturally suggests many interesting families of MNNs which can be interpreted as the manifold analog of various popular GNNs. We then propose a method for implementing MFCNs on high-dimensional point clouds that relies on approximating the manifold by a sparse graph. We prove that our method is consistent in the sense that it converges to a continuum limit as the number of data points tends to infinity.