An elementary method to compute equivariant convolutional kernels on homogeneous spaces for geometric deep learning

03/13/2023
by   Vincent Knibbeler, et al.
0

We develop an elementary method to compute spaces of equivariant maps from a homogeneous space of a Lie group to a module of this group. The Lie group is not required to be compact. More generally we study spaces of invariant sections in homogeneous vector bundles, and take a special interest in the case where the fibres are algebras. This latter case has a natural global algebra structure. We classify the resulting automorphic algebras for the case where the homogeneous space has compact stabilisers. This work has applications in the theoretical development of geometric deep learning and also in the theory of automorphic Lie algebras.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
05/12/2021

Homogeneous vector bundles and G-equivariant convolutional neural networks

G-equivariant convolutional neural networks (GCNNs) is a geometric deep ...
research
09/29/2022

Equivariant maps from invariant functions

In equivariant machine learning the idea is to restrict the learning to ...
research
01/28/2022

High-order integrators for Lagrangian systems on homogeneous spaces via nonholonomic mechanics

In this paper, high-order numerical integrators on homogeneous spaces wi...
research
08/01/2023

Geometry preserving numerical methods for physical systems with finite-dimensional Lie algebras

In this paper we propose a geometric integrator to numerically approxima...
research
06/16/2022

Unified Fourier-based Kernel and Nonlinearity Design for Equivariant Networks on Homogeneous Spaces

We introduce a unified framework for group equivariant networks on homog...
research
08/27/2020

Bi-invariant Two-Sample Tests in Lie Groups for Shape Analysis

We propose generalizations of the Hotelling's T^2 statistic and the Bhat...

Please sign up or login with your details

Forgot password? Click here to reset