Frequency Selective Neural Networks as a Foundation Architecture for Time Series Learning

2026-08-28

AI Summary

本文提出频率选择性神经网络（FSNN），通过嵌入高级信号处理数学方法解决频谱纠缠问题，提高时间序列学习的物理可解释性和预测性能。

Abstract

Time-series data across physical and biological domains are fundamentally driven by complex, non-stationary oscillatory modes. While deep learning models, such as Convolutional Neural Networks (CNNs), Recurrent Neural Networks, and Transformers, have dominated sequential analysis, they remain fundamentally "spectral-blind". By mapping continuous physical waves into unconstrained spatial or discrete token spaces, these architectures suffer from severe spectral entanglement, acting as opaque black boxes that decouple predictive accuracy from physical reality. In this paper, we introduce the Frequency Selective Neural Network (FSNN), pioneering a foundation architecture guaranteeing physical interpretability without sacrificing expressive power of deep learning. FSNN addresses spectral entanglement by explicitly embedding the rigorous mathematics of advanced signal processing into its neural topology. Through a fully differentiable Wiener-like filter bank optimized via complex-domain backpropagation, FSNN autonomously discovers and isolates the precise physical modes of a given task. Extensive evaluations demonstrate that FSNN establishes state-of-the-art predictive performance, achieving $77.0\%$ average accuracy on the standard 10 multivariate UEA datasets and leading across all major metrics on the highly imbalanced PTB-XL clinical ECG benchmark. Crucially, in contrast to yielding abstract feature maps, FSNN converges directly on physically meaningful frequency bands, such as isolating the cardiac QRS complex, providing a highly scalable, interpretable paradigm for robust pattern recognition in complex temporal domains. Our code is available at: https://github.com/ad6174hhhh/FSNN.

Problem

• Spectral Entanglement
• Time Series Learning
• Physical Interpretability

Innovation

• Frequency Selective Neural Network
• Spectral Entanglement
• Physical Interpretability
• Wiener-like Filter Bank
• Complex-domain Backpropagation