报告人：游子琳 博士生（上海交通大学）

时间：2026年01月09日 14:30-

地址：理科楼LA104

摘要：We present a multi-scale Fourier neural operator (MscaleFNO) designed to reduce the spectral bias of the FNO in learning the mapping between highly oscillatory functions, particularly focusing on the nonlinear mapping between the coefficient of the Helmholtz equation and its solution. The proposed MscaleFNO architecture employs parallel normal FNOs with scaled input of the function and the spatial variable, enabling effective capture of various high-frequency components in the mapping's image. Numerical experiments demonstrate that MscaleFNO achieves substantial improvements in solving wave scattering problems in the high-frequency regime compared to the normal FNO, while maintaining a similar parameter number.

邀请人：数学研究中心

欢迎广大师生积极参与！