CHIME: Conditional Hallucination and Integrated Multi-scale Enhancement for Time Series Diffusion Model
This work addresses time series generation problems for applications requiring accurate modeling across entities and long-time scales, representing an incremental improvement over existing diffusion model approaches.
The paper tackles challenges in multi-scale feature alignment and generative capabilities for time series diffusion models by proposing CHIME, a framework that uses multi-scale decomposition and a feature hallucination module, achieving state-of-the-art performance and excellent generative generalization in few-shot scenarios on real-world datasets.
The denoising diffusion probabilistic model has become a mainstream generative model, achieving significant success in various computer vision tasks. Recently, there has been initial exploration of applying diffusion models to time series tasks. However, existing studies still face challenges in multi-scale feature alignment and generative capabilities across different entities and long-time scales. In this paper, we propose CHIME, a conditional hallucination and integrated multi-scale enhancement framework for time series diffusion models. By employing multi-scale decomposition and integration, CHIME captures the decomposed features of time series, achieving in-domain distribution alignment between generated and original samples. In addition, we introduce a feature hallucination module in the conditional denoising process, enabling the temporal features transfer across long-time scales. Experimental results on publicly available real-world datasets demonstrate that CHIME achieves state-of-the-art performance and exhibits excellent generative generalization capabilities in few-shot scenarios.