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

2026-03-30

AI Summary

This work addresses the limitations of fixed input lengths and insufficient multi-scale modeling in time series forecasting by proposing a novel approach that integrates hierarchical multi-resolution trend decomposition, an adaptive embedding mechanism, and multi-scale conditional diffusion generation. For the first time, a dual-image representation is incorporated into a multi-scale conditional diffusion model, enabling efficient multi-resolution modeling of variable-length sequences. Experimental results on four real-world datasets demonstrate that the proposed method significantly outperforms state-of-the-art models such as CSDI and Informer, achieving average reductions of 6%–10% in both MAE and RMSE metrics.

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.

Problem

• time series forecasting
• fixed-length inputs
• multi-scale modeling
• variable-length inputs

Innovation

• multi-resolution
• conditional diffusion
• adaptive embedding
• time series forecasting
• hierarchical decomposition