AI RESEARCH
Multi-Scale Separable Fourier Neural Networks for Solving High-Frequency PDEs
arXiv CS.LG
•
ArXi:2605.31027v1 Announce Type: new We propose a novel neural network architecture, termed Multi-Scale Separable Fourier Neural Networks (MS-SFNN), for the accurate and efficient solution of linear and nonlinear high-frequency partial differential equations (PDEs). MS-SFNN exploits a separable representation: given a $d$-dimensional input, it employs $d$ independent subnetworks -- each acting on a single coordinate -- and constructs basis functions via element-wise multiplication of their outputs.