Sub-Riemannian Landmark Matching and its interpretation as residual neural networks

arXiv2204.09351MaRDI QIDQ6505779FDOQ6505779

Abstract: The problem of finding a time-dependent vector field which warps an initial set of points to a target set is common in shape analysis. It is an example of a problem in the diffeomorphic shape matching regime and can be thought of as a type of spatial discretization of the more general LDDMM framework as it transforms into a finite-dimensional Hamiltonian system. In this paper, we consider landmark matching modified by restricting the set of available vector fields in the sense that vector fields are parametrized by a set of controls. We determine the geometric setting of the problem, referred to as sub-Riemannian landmark matching, and derive equations of motions for the controls. We provide two computational algorithms and demonstrate them in numerical examples. In particular, the experiments highlight the importance of the regularization term. The reason why this is of interest is that sub-Riemannian landmark matching is demonstrated to have connections with neural networks, in particular the interpretation of residual neural networks as time discretizations of continuous control problems. It allows shape analysis practitioners to think about neural networks in terms of shape analysis, thereby providing a bridge between the two fields.

Mathematics Subject Classification ID

Artificial neural networks and deep learning (68T07) Sub-Riemannian geometry (53C17) Differential geometric methods (tensors, connections, symplectic, Poisson, contact, Riemannian, nonholonomic, etc.) for problems in mechanics (70G45) Hamiltonian and Lagrangian mechanics (70H99)

This page was built for publication: Sub-Riemannian Landmark Matching and its interpretation as residual neural networks

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6505779)