Riemannian Score-Based Generative Modeling

02/06/2022
by   Valentin De Bortoli, et al.
5

Score-based generative models (SGMs) are a novel class of generative models demonstrating remarkable empirical performance. One uses a diffusion to add progressively Gaussian noise to the data, while the generative model is a "denoising" process obtained by approximating the time-reversal of this "noising" diffusion. However, current SGMs make the underlying assumption that the data is supported on a Euclidean manifold with flat geometry. This prevents the use of these models for applications in robotics, geoscience or protein modeling which rely on distributions defined on Riemannian manifolds. To overcome this issue, we introduce Riemannian Score-based Generative Models (RSGMs) which extend current SGMs to the setting of compact Riemannian manifolds. We illustrate our approach with earth and climate science data and show how RSGMs can be accelerated by solving a Schrödinger bridge problem on manifolds.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
07/07/2022

Riemannian Diffusion Schrödinger Bridge

Score-based generative models exhibit state of the art performance on de...
research
03/17/2022

Visualizing Riemannian data with Rie-SNE

Faithful visualizations of data residing on manifolds must take the unde...
research
05/24/2023

Manifold Diffusion Fields

We present Manifold Diffusion Fields (MDF), an approach to learn generat...
research
04/11/2023

Diffusion Models for Constrained Domains

Denoising diffusion models are a recent class of generative models which...
research
07/11/2023

Metropolis Sampling for Constrained Diffusion Models

Denoising diffusion models have recently emerged as the predominant para...
research
03/25/2020

A diffusion approach to Stein's method on Riemannian manifolds

We detail an approach to develop Stein's method for bounding integral me...
research
08/20/2019

Expected path length on random manifolds

Manifold learning seeks a low dimensional representation that faithfully...

Please sign up or login with your details

Forgot password? Click here to reset