🤖 AI Summary
Existing time-series generation methods lack interpretable decomposition mechanisms, hindering faithful modeling of meaningful trend and seasonal patterns. To address this, we propose Seasonal-Trend Diffusion (STDiffusion), the first framework to deeply integrate learnable sequence decomposition with diffusion modeling. STDiffusion explicitly parameterizes the trend component via MLPs and captures multi-scale seasonality through adaptive wavelet distillation; it further introduces a component correction mechanism to enforce consistency and decoupling between trend and seasonal components. This design significantly enhances both interpretability across multiple resolutions and internal coherence of generated series. Evaluated on eight real-world datasets, STDiffusion achieves state-of-the-art performance across all generation benchmarks. Moreover, it demonstrates strong robustness and generalization in multi-window, long-horizon forecasting tasks.
📝 Abstract
Time series generation focuses on modeling the underlying data distribution and resampling to produce authentic time series data. Key components, such as trend and seasonality, drive temporal fluctuations, yet many existing approaches fail to employ interpretative decomposition methods, limiting their ability to synthesize meaningful trend and seasonal patterns. To address this gap, we introduce Seasonal-Trend Diffusion (STDiffusion), a novel framework for multivariate time series generation that integrates diffusion probabilistic models with advanced learnable series decomposition techniques, enhancing the interpretability of the generation process. Our approach separates the trend and seasonal learning into distinct blocks: a Multi-Layer Perceptron (MLP) structure captures the trend, while adaptive wavelet distillation facilitates effective multi-resolution learning of seasonal components. This decomposition improves the interpretability of the model on multiple scales. In addition, we designed a comprehensive correction mechanism aimed at ensuring that the generated components exhibit a high degree of internal consistency and preserve meaningful interrelationships with one another. Our empirical studies on eight real-world datasets demonstrate that STDiffusion achieves state-of-the-art performance in time series generation tasks. Furthermore, we extend the model's application to multi-window long-sequence time series generation, which delivered reliable results and highlighted its robustness and versatility.