Fourier-mixed window attention for efficient and robust long sequence time-series forecasting
Year: 2025
Type: article
Abstract: We study a fast local-global window-based attention method to accelerate Informer for long sequence time-series forecasting (LSTF) in a robust manner. While window attention being local is a considerable computational saving, it lacks the ability to capture global token information which is compensated by a subsequent Fourier transform block. Our method, named FWin, does not rely on query sparsity hypothesis and an empirical approximation underlying the ProbSparse attention of Informer. Experiments on univariate and multivariate datasets show that FWin transformers improve the overall prediction accuracies of Informer while accelerating its inference speeds by 1.6 to 2 times. On strongly non-stationary data (power grid and dengue disease data), FWin outperforms Informer and recent SOTAs thereby demonstrating its superior robustness. We give mathematical definition of FWin attention, and prove its equivalency to the canonical full attention under the block diagonal invertibility (BDI) condition of the attention matrix. The BDI is verified to hold with high probability on benchmark datasets experimentally.
Authors: Nhat Thanh Tran, Jack Xin
Institution: University of California, Irvine
Language: English
Subfield: Signal Processing
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $1,418
Funders: National Science Foundation, Qualcomm
Awards: Collaborative Research: ATD: Fast Algorithms and Novel Continuous-depth Graph Neural Networks for Threat Detection, Computational and Mathematical Studies of Compression and Distillation Methods for Deep Neural Networks and Applications, FRG: Collaborative Research: Robust, Efficient, and Private Deep Learning Algorithms
OpenAlex ID: W4410508667