北京大学定量生物学中心
学术报告
题 目: Advancing molecular physics with deep learning
报告人: Prof. Frank Noé
The AI4Science Group, Freie Universität Berlin
Microsoft Research (MSR) AI4Science
时 间: 6月6日（周四）11:00-12:00
地 点: 吕志和楼B101
主持人: 宋晨 研究员
摘要:
Molecular simulation may ideally serve as a "computational laboratory", with is ability to observe both structure and dynamics at high resolution and to simulate molecules that are difficult to synthesize. However, it also suffers from fundamental limitations, in particular in the accurate modeling of molecules and in the efficient computation of experimental observables. By leveraging the latest developments in machine learning, we can advance molecular simulation algorithms to make significant progress at these fronts without sacrificing physics.
In this talk, I will give an overview over our work on the highly accurate computation of quantum states with deep fermionic neural networks and Quantum Monte Carlo and addressing the many-body sampling problem using generative deep learning.
Frank has done his undergraduate degrees in electrical engineering (Stuttgart) and computer science (Cork, Ireland) and his PhD in biophysics at University of Heidelberg in 2006. Shortly after he started a research group at FU Berlin and was promoted to tenured professor in 2013. Frank has received two ERC grants, the early career award in theoretical Chemistry of the American Chemical Society. Since 2019 he is in the list ISI highly cited researchers (top 1% of their field), and a member of the Berlin-Brandenburg Academy of Sciences. Since 2022 Frank is Partner Research Manager at Microsoft Research AI for Science and is leading a new research team in Berlin, Germany. Frank's research focus is on developing ML + AI methods for the natural sciences, in particular molecular physics.
