MR-ImagenTime: Multi-Resolution Time Series Generation through Dual Image Representations
This work improves time series forecasting for domains like finance or healthcare, but it appears incremental as it builds on existing diffusion and decomposition methods.
The paper tackled the problem of time series forecasting by addressing fixed-length inputs and multi-scale modeling, proposing MR-CDM which significantly outperformed state-of-the-art baselines by reducing MAE and RMSE by approximately 6-10%.
Time series forecasting is vital across many domains, yet existing models struggle with fixed-length inputs and inadequate multi-scale modeling. We propose MR-CDM, a framework combining hierarchical multi-resolution trend decomposition, an adaptive embedding mechanism for variable-length inputs, and a multi-scale conditional diffusion process. Evaluations on four real-world datasets demonstrate that MR-CDM significantly outperforms state-of-the-art baselines (e.g., CSDI, Informer), reducing MAE and RMSE by approximately 6-10 to a certain degree.