From the Greene–Wu Convolution to Gradient Estimation over Riemannian Manifolds

08/17/2021
by   Tianyu Wang, et al.
0

Over a complete Riemannian manifold of finite dimension, Greene and Wu introduced a convolution, known as Greene-Wu (GW) convolution. In this paper, we study properties of the GW convolution and apply it to non-Euclidean machine learning problems. In particular, we derive a new formula for how the curvature of the space would affect the curvature of the function through the GW convolution. Also, following the study of the GW convolution, a new method for gradient estimation over Riemannian manifolds is introduced.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
12/24/2019

Constant index expectation curvature for graphs or Riemannian manifolds

An integral geometric curvature is defined as the index expectation K(x)...
research
04/19/2012

Learning in Riemannian Orbifolds

Learning in Riemannian orbifolds is motivated by existing machine learni...
research
02/17/2023

Minimizing Dynamic Regret on Geodesic Metric Spaces

In this paper, we consider the sequential decision problem where the goa...
research
03/08/2018

Generalized partially linear models on Riemannian manifolds

The generalized partially linear models on Riemannian manifolds are intr...
research
12/02/2019

More on Poincare-Hopf and Gauss-Bonnet

We illustrate connections between differential geometry on finite simple...
research
11/06/2019

Optimal Design of Experiments on Riemannian Manifolds

Traditional optimal design of experiment theory is developed on Euclidea...
research
07/22/2019

Barycentric interpolation on Riemannian and semi-Riemannian spaces

Interpolation of data represented in curvilinear coordinates and possibl...

Please sign up or login with your details

Forgot password? Click here to reset