报告时间：2025年2月27日19:30
报告地点：腾讯会议：935-904-097
报告人：蒲俊才 北京应用物理与计算数学研究所
报告摘要：
Deep learning for solving PDE has become a research focus in various fields. We first briefly introduce the development history and basic knowledge of deep learning. The NLS equation is a classical integrable equation which contains plenty of significant properties and occurs in many physical areas. Recently, the PINN is proposed to uncover the equation’s dynamical behaviors from spatiotemporal data directly. Compared with traditional neural networks, this method can obtain remarkably accurate solution with extraordinarily less data. We utilize the PINN to solve the soliton solutions, breather solution, and rogue wave solutions of the NLS equation. In particular, the dynamical behaviors and error analysis about the one-order and two-order rogue waves of nonlinear integrable equations are revealed by the deep learning method for the first time.
报告人简介：
蒲俊才，北京应用物理与计算数学研究所博士后。2023年6月博士毕业于华东师范大学数学科学学院应用数学专业，从事可积系统以及机器学习方面的研究。在国内外知名期刊J. Comput. Phys.、Phys. D、J. Math. Phys.等上发表SCI论文10余篇。主持中国博士后科学基金特别资助1项，中国博士后科学基金面上资助1项。