Sawtooth Sampling for Time Series Denoising Diffusion Implicit Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Oppel, Heiko, Spilz, Andreas, Munz, Michael
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908677054660608
author Oppel, Heiko
Spilz, Andreas
Munz, Michael
author_facet Oppel, Heiko
Spilz, Andreas
Munz, Michael
contents Denoising Diffusion Probabilistic Models (DDPMs) can generate synthetic timeseries data to help improve the performance of a classifier, but their sampling process is computationally expensive. We address this by combining implicit diffusion models with a novel Sawtooth Sampler that accelerates the reverse process and can be applied to any pretrained diffusion model. Our approach achieves a 30 times speed-up over the standard baseline while also enhancing the quality of the generated sequences for classification tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2511_21320
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Sawtooth Sampling for Time Series Denoising Diffusion Implicit Models
Oppel, Heiko
Spilz, Andreas
Munz, Michael
Machine Learning
Denoising Diffusion Probabilistic Models (DDPMs) can generate synthetic timeseries data to help improve the performance of a classifier, but their sampling process is computationally expensive. We address this by combining implicit diffusion models with a novel Sawtooth Sampler that accelerates the reverse process and can be applied to any pretrained diffusion model. Our approach achieves a 30 times speed-up over the standard baseline while also enhancing the quality of the generated sequences for classification tasks.
title Sawtooth Sampling for Time Series Denoising Diffusion Implicit Models
topic Machine Learning
url https://arxiv.org/abs/2511.21320