Log In Sign Up

Fully-Connected Network on Noncompact Symmetric Space and Ridgelet Transform based on Helgason-Fourier Analysis

by   Sho Sonoda, et al.

Neural network on Riemannian symmetric space such as hyperbolic space and the manifold of symmetric positive definite (SPD) matrices is an emerging subject of research in geometric deep learning. Based on the well-established framework of the Helgason-Fourier transform on the noncompact symmetric space, we present a fully-connected network and its associated ridgelet transform on the noncompact symmetric space, covering the hyperbolic neural network (HNN) and the SPDNet as special cases. The ridgelet transform is an analysis operator of a depth-2 continuous network spanned by neurons, namely, it maps an arbitrary given function to the weights of a network. Thanks to the coordinate-free reformulation, the role of nonlinear activation functions is revealed to be a wavelet function, and the reconstruction formula directly yields the universality of the proposed networks.


page 1

page 2

page 3

page 4


Universality of group convolutional neural networks based on ridgelet analysis on groups

We investigate the approximation property of group convolutional neural ...

A Simple Riemannian Manifold Network for Image Set Classification

In the domain of image-set based classification, a considerable advance ...

An exploration of parameter redundancy in deep networks with circulant projections

We explore the redundancy of parameters in deep neural networks by repla...

Gaussian distributions on Riemannian symmetric spaces in the large N limit

We consider Gaussian distributions on certain Riemannian symmetric space...

A Coordinated View of Cyberspace

Cyberspace is an online world created by growing network of computing an...

Circular-Symmetric Correlation Layer based on FFT

Despite the vast success of standard planar convolutional neural network...

Bit Complexity of Computing Solutions for Symmetric Hyperbolic Systems of PDEs with Guaranteed Precision

We establish upper bounds of bit complexity of computing solution operat...