English

Sawtooth Sampling for Time Series Denoising Diffusion Implicit Models

Machine Learning 2025-11-27 v1

Abstract

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.

Keywords

Cite

@article{arxiv.2511.21320,
  title  = {Sawtooth Sampling for Time Series Denoising Diffusion Implicit Models},
  author = {Heiko Oppel and Andreas Spilz and Michael Munz},
  journal= {arXiv preprint arXiv:2511.21320},
  year   = {2025}
}
R2 v1 2026-07-01T07:56:04.162Z