Speaker: Prof. Zhiping Mao（毛志平教授）

Time: 9:00-10:00, 20 March 2023 (Monday) (Beijing time)

Venue: A103,Lijiao Building

Tencent Meeting ID: 679-222-969

Abstract

In the talk, I shall present some numerical results concerning the forward and inverse problems of PDEs with sharp solutions using deep neural networks and deep operator neural networks-based methods. Firstly, we solve the inverse and forward problems of PDEs with sharp solutions using PINNs with residual/gradient-based adaptive sampling. Secondly, we develop several types of neural networks architecture based on deep neural networks and deep operator neural networks for solving multiscale multiphysics problems as well as sharp interface problems. In particular, we combine the pretrained deep neural operator and PINNs to efficiently solving the problems whose solutions have sharpness.

About Prof. Mao

毛志平，厦门大学数学科学学院教授，2009年本科毕业于重庆大学，2015年博士毕业于厦门大学计算数学专业，国家高层次青年人才，2015年10月至2020年9月在美国布朗大学应用数学系从事博士后研究。毛志平博士主要从事谱方法以及机器学习方面的研究，其目前在SIREV, JCP，SISC，SINUM、CMAME等国际高水平杂志上发表论文20余篇。