MR-ImagenTime: Multi-Resolution Time Series Generation through Dual Image Representations

Abstract

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.

Figures (6)

• Figure 1: Multi-Resolution Time Series Example
• Figure 2: MR-CDM Model Architecture
• Figure 3: The Prediction Result Based on Our MR-CDM Model
• Figure 4: The Prediction Result Based on CSDI Model
• Figure 5: Multi-Step Prediction Performance Comparison