IPatch: A Multi-Resolution Transformer Architecture for Robust Time-Series Forecasting
This work addresses the problem of robust time-series forecasting for applications requiring handling of both short-term fluctuations and long-range dependencies, representing an incremental improvement over existing Transformer-based methods.
The paper tackled the challenge of accurately forecasting multivariate time series by proposing IPatch, a multi-resolution Transformer architecture that integrates point-wise and patch-wise tokens to capture both fine-grained details and broader dependencies, resulting in improved accuracy, robustness to noise, and generalization across 7 benchmark datasets.
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.