TimeDP: Learning to Generate Multi-Domain Time Series with Domain Prompts

TL;DR

TimeDP presents a novel multi-domain time series diffusion model that learns a fixed set of time-series prototypes and uses a prototype-assignment module to form domain prompts. These prompts condition a diffusion backbone to generate time series across multiple domains without explicit domain labels, enabling strong in-domain quality and robust unseen-domain synthesis via few-shot prompts. The approach achieves state-of-the-art generation performance on 12 real-world datasets and demonstrates zero-shot transfer to unseen domains, with ablations confirming the critical roles of PAM and domain prompts. This framework offers a scalable, label-free mechanism to bridge diverse time-series domains for data augmentation, privacy, and scenario simulation in practice.

Abstract

Time series generation models are crucial for applications like data augmentation and privacy preservation. Most existing time series generation models are typically designed to generate data from one specified domain. While leveraging data from other domain for better generalization is proved to work in other application areas, this approach remains challenging for time series modeling due to the large divergence in patterns among different real world time series categories. In this paper, we propose a multi-domain time series diffusion model with domain prompts, named TimeDP. In TimeDP, we utilize a time series semantic prototype module which defines time series prototypes to represent time series basis, each prototype vector serving as "word" representing some elementary time series feature. A prototype assignment module is applied to extract the extract domain specific prototype weights, for learning domain prompts as generation condition. During sampling, we extract "domain prompt" with few-shot samples from the target domain and use the domain prompts as condition to generate time series samples. Experiments demonstrate that our method outperforms baselines to provide the state-of-the-art in-domain generation quality and strong unseen domain generation capability.

Figures (7)

• Figure 1: Overview of TimeDP model.
• Figure 2: Semantic visualization of time series prototypes.
• Figure 3: Heatmap of Domain Prompts.
• Figure 4: T-SNE visualization of domain prompts. Domain prompt generated for each dataset are marked with the same color.
• Figure 5: T-SNE visualization of domain prompts. Datasets in the same domain are marked with the same color.