FlexTSF: A Flexible Forecasting Model for Time Series with Variable Regularities
This work addresses forecasting for time series with variable regularities, a common issue in real-world applications due to diverse sensing devices, but it is incremental as it builds on existing continuous-time and patching methods.
The authors tackled the problem of forecasting time series with irregular temporal structures, which is challenging for existing models that assume regular sampling or rely on imputation. They introduced FlexTSF, a flexible forecasting model that significantly outperformed existing models in experiments on 16 datasets across classic forecasting, zero-shot generalization, and low-resource fine-tuning scenarios.
Forecasting time series with irregular temporal structures remains challenging for universal pre-trained models. Existing approaches often assume regular sampling or depend heavily on imputation, limiting their applicability in real-world scenarios where irregularities are prevalent due to diverse sensing devices and recording practices. We introduce FlexTSF, a flexible forecasting model specifically designed for time series data with variable temporal regularities. At its foundation lies the IVP Patcher, a continuous-time patching module leveraging Initial Value Problems (IVPs) to inherently support uneven time intervals, variable sequence lengths, and missing values. FlexTSF employs a decoder-only architecture that integrates normalized timestamp inputs and domain-specific statistics through a specialized causal self-attention mechanism, enabling adaptability across domains. Extensive experiments on 16 datasets demonstrate FlexTSF's effectiveness, significantly outperforming existing models in classic forecasting scenarios, zero-shot generalization, and low-resource fine-tuning conditions. Ablation studies confirm the contributions of each design component and the advantage of not relying on predefined fixed patch lengths.