IPatch: A Multi-Resolution Transformer Architecture for Robust Time-Series Forecasting

TL;DR

IPatch introduces a multi-resolution Transformer model that combines point-wise and patch-wise representations to enhance the accuracy, robustness, and scalability of multivariate time-series forecasting.

Contribution

It is the first to integrate multi-resolution representations within a Transformer for time-series forecasting, improving performance over single-resolution models.

Findings

Consistently outperforms baseline models on 7 benchmark datasets.

Enhances robustness to noise and complex temporal patterns.

Improves forecasting accuracy across various horizons.

Abstract

Accurate forecasting of multivariate time series remains challenging due to the need to capture both short-term fluctuations and long-range temporal dependencies. Transformer-based models have emerged as a powerful approach, but their performance depends critically on the representation of temporal data. Traditional point-wise representations preserve individual time-step information, enabling fine-grained modeling, yet they tend to be computationally expensive and less effective at modeling broader contextual dependencies, limiting their scalability to long sequences. Patch-wise representations aggregate consecutive steps into compact tokens to improve efficiency and model local temporal dynamics, but they often discard fine-grained temporal details that are critical for accurate predictions in volatile or complex time series. We propose IPatch, a multi-resolution Transformer architecture that integrates both point-wise and patch-wise tokens, modeling temporal information at multiple resolutions. Experiments on 7 benchmark datasets demonstrate that IPatch consistently improves forecasting accuracy, robustness to noise, and generalization across various prediction horizons compared to single-representation baselines.