发表机构
Coginfinite Technology; University of Virginia; University of Hong Kong（科无限科技; 弗吉尼亚大学; 香港大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究针对现有深度学习模型处理时间序列时的光谱纠缠问题，提出频率选择神经网络（FSNN），通过嵌入信号处理数学实现物理可解释性，在多变量UEA数据集和PTB-XL基准测试中取得最优性能。
AI 中文摘要
物理和生物领域的时间序列数据根本上由复杂、非平稳的振荡模式驱动。尽管卷积神经网络（CNN）、循环神经网络和Transformer等深度学习模型在序列分析中占据主导地位，但它们本质上是“光谱盲”的。这些架构将连续物理波映射到无约束的空间或离散令牌空间，会遭受严重的光谱纠缠，成为不透明的黑箱，使预测精度与物理现实脱节。本文提出频率选择神经网络（FSNN），开创了在不牺牲深度学习表达能力的前提下保证物理可解释性的基础架构。FSNN通过将高级信号处理的严谨数学明确嵌入其神经拓扑结构来解决光谱纠缠问题，借助通过复域反向传播优化的完全可微类维纳滤波器组，FSNN能自主发现并隔离给定任务的精确物理模式。大量评估表明，FSNN达到了最先进的预测性能：在标准10个多变量UEA数据集上平均准确率达77.0%，在高度不平衡的PTB-XL临床心电图基准测试的所有主要指标中均领先。关键的是，与生成抽象特征图不同，FSNN会直接收敛到具有物理意义的频带，例如隔离心脏QRS复合波，为复杂时间域的鲁棒模式识别提供了高度可扩展、可解释的范式。我们的代码可在以下网址获取：this https URL。
英文摘要
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.
