报告题目：Deep Learning for Inverse Problems in Physics
报告人：周凯，法兰克福高等研究院
报告摘要：Inverse Problems occur in almost all research areas. Due to the indirect noisy observation or even 'ill-posedness', it's usually challenging to handle the inverse problem. In this talk I will introduce some projects that utilizing deep learning techniques for solving inverse problems in physics, especially in the context of basic research for the exploration of matter in high energy nuclear physics. Specifically I will talk about early time physics identification from heavy-ion collisions, heavy-quark potential inference from lattice measurements, spectral function reconstruction from correlator, learning Neutron Star Equation of State from observatory, and microscopic interaction modelling based on density estimation for many-body system.
报告人简介：Dr. Kai Zhou received his B.Sc. degree in Physics from Xi'an Jiaotong University in 2009, and his PhD degree in physics from Tsinghua University (Superviser: Prof. Pengfei Zhuang) with "Wu You Xun" Honors in 2014 . Afterwards he went to Goethe University for his Postdoctoral research in the Institute for Theoretical Physics (ITP). Since 2017, he joined FIAS as Research Fellow and lead the group "Deepthinkers" focusing on Deep Learning (DL) for physics and beyond, and since 2021 he became fellow at FIAS. Dr. Zhou has a very broad interest in physics and AI/DL application in different fields, particularly developing data-driven and physics-informed deep learning methods for physics research.
