Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Young, James Matthew, Cordero-Encinar, Paula, Reich, Sebastian, Duncan, Andrew, Akyildiz, O. Deniz
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2601.21951
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915990316515328
author Young, James Matthew
Cordero-Encinar, Paula
Reich, Sebastian
Duncan, Andrew
Akyildiz, O. Deniz
author_facet Young, James Matthew
Cordero-Encinar, Paula
Reich, Sebastian
Duncan, Andrew
Akyildiz, O. Deniz
contents We develop diffusion-based samplers for target distributions known up to a normalising constant. To this end, we rely on the well-known diffusion path that smoothly interpolates between a simple base distribution and the target, popularised by diffusion models. We tackle the score estimation problem by developing an efficient sequential Monte Carlo sampler that evolves auxiliary variables from conditional distributions along the path, providing principled score and density estimates for time-varying distributions. To control the variance of score estimates, we further propose practical control variate schedules that incur minimal overhead. We adapt this general framework to paths induced by the Ornstein-Uhlenbeck (OU) time-reversal process, stochastic interpolants, and diffusion annealed Langevin dynamics, outlining their trade-offs. Finally, we provide theoretical guarantees and empirically demonstrate the effectiveness of our method on several synthetic and real-world datasets.
format Preprint
id arxiv_https___arxiv_org_abs_2601_21951
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Diffusion Path Samplers via Sequential Monte Carlo
Young, James Matthew
Cordero-Encinar, Paula
Reich, Sebastian
Duncan, Andrew
Akyildiz, O. Deniz
Machine Learning
Computation
We develop diffusion-based samplers for target distributions known up to a normalising constant. To this end, we rely on the well-known diffusion path that smoothly interpolates between a simple base distribution and the target, popularised by diffusion models. We tackle the score estimation problem by developing an efficient sequential Monte Carlo sampler that evolves auxiliary variables from conditional distributions along the path, providing principled score and density estimates for time-varying distributions. To control the variance of score estimates, we further propose practical control variate schedules that incur minimal overhead. We adapt this general framework to paths induced by the Ornstein-Uhlenbeck (OU) time-reversal process, stochastic interpolants, and diffusion annealed Langevin dynamics, outlining their trade-offs. Finally, we provide theoretical guarantees and empirically demonstrate the effectiveness of our method on several synthetic and real-world datasets.
title Diffusion Path Samplers via Sequential Monte Carlo
topic Machine Learning
Computation
url https://arxiv.org/abs/2601.21951