Random-Bridges as Stochastic Transports for Generative Models

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Goria, Stefano, Mengütürk, Levent A., Mengütürk, Murat C., Sesen, Berkan
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866917383430471680
author Goria, Stefano
Mengütürk, Levent A.
Mengütürk, Murat C.
Sesen, Berkan
author_facet Goria, Stefano
Mengütürk, Levent A.
Mengütürk, Murat C.
Sesen, Berkan
contents This paper motivates the use of random-bridges -- stochastic processes conditioned to take target distributions at fixed timepoints -- in the realm of generative modelling. Herein, random-bridges can act as stochastic transports between two probability distributions when appropriately initialized, and can display either Markovian or non-Markovian, and either continuous, discontinuous or hybrid patterns depending on the driving process. We show how one can start from general probabilistic statements and then branch out into specific representations for learning and simulation algorithms in terms of information processing. Our empirical results, built on Gaussian random bridges, produce high-quality samples in significantly fewer steps compared to traditional approaches, while achieving competitive Frechet inception distance scores. Our analysis provides evidence that the proposed framework is computationally cheap and suitable for high-speed generation tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2512_14190
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Random-Bridges as Stochastic Transports for Generative Models
Goria, Stefano
Mengütürk, Levent A.
Mengütürk, Murat C.
Sesen, Berkan
Machine Learning
Probability
This paper motivates the use of random-bridges -- stochastic processes conditioned to take target distributions at fixed timepoints -- in the realm of generative modelling. Herein, random-bridges can act as stochastic transports between two probability distributions when appropriately initialized, and can display either Markovian or non-Markovian, and either continuous, discontinuous or hybrid patterns depending on the driving process. We show how one can start from general probabilistic statements and then branch out into specific representations for learning and simulation algorithms in terms of information processing. Our empirical results, built on Gaussian random bridges, produce high-quality samples in significantly fewer steps compared to traditional approaches, while achieving competitive Frechet inception distance scores. Our analysis provides evidence that the proposed framework is computationally cheap and suitable for high-speed generation tasks.
title Random-Bridges as Stochastic Transports for Generative Models
topic Machine Learning
Probability
url https://arxiv.org/abs/2512.14190