Bridging discrete and continuous state spaces: Exploring the Ehrenfest process in time-continuous diffusion models

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Winkler, Ludwig, Richter, Lorenz, Opper, Manfred
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866913342809964544
author Winkler, Ludwig
Richter, Lorenz
Opper, Manfred
author_facet Winkler, Ludwig
Richter, Lorenz
Opper, Manfred
contents Generative modeling via stochastic processes has led to remarkable empirical results as well as to recent advances in their theoretical understanding. In principle, both space and time of the processes can be discrete or continuous. In this work, we study time-continuous Markov jump processes on discrete state spaces and investigate their correspondence to state-continuous diffusion processes given by SDEs. In particular, we revisit the $\textit{Ehrenfest process}$, which converges to an Ornstein-Uhlenbeck process in the infinite state space limit. Likewise, we can show that the time-reversal of the Ehrenfest process converges to the time-reversed Ornstein-Uhlenbeck process. This observation bridges discrete and continuous state spaces and allows to carry over methods from one to the respective other setting. Additionally, we suggest an algorithm for training the time-reversal of Markov jump processes which relies on conditional expectations and can thus be directly related to denoising score matching. We demonstrate our methods in multiple convincing numerical experiments.
format Preprint
id arxiv_https___arxiv_org_abs_2405_03549
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Bridging discrete and continuous state spaces: Exploring the Ehrenfest process in time-continuous diffusion models
Winkler, Ludwig
Richter, Lorenz
Opper, Manfred
Machine Learning
Dynamical Systems
Probability
Generative modeling via stochastic processes has led to remarkable empirical results as well as to recent advances in their theoretical understanding. In principle, both space and time of the processes can be discrete or continuous. In this work, we study time-continuous Markov jump processes on discrete state spaces and investigate their correspondence to state-continuous diffusion processes given by SDEs. In particular, we revisit the $\textit{Ehrenfest process}$, which converges to an Ornstein-Uhlenbeck process in the infinite state space limit. Likewise, we can show that the time-reversal of the Ehrenfest process converges to the time-reversed Ornstein-Uhlenbeck process. This observation bridges discrete and continuous state spaces and allows to carry over methods from one to the respective other setting. Additionally, we suggest an algorithm for training the time-reversal of Markov jump processes which relies on conditional expectations and can thus be directly related to denoising score matching. We demonstrate our methods in multiple convincing numerical experiments.
title Bridging discrete and continuous state spaces: Exploring the Ehrenfest process in time-continuous diffusion models
topic Machine Learning
Dynamical Systems
Probability
url https://arxiv.org/abs/2405.03549