Sunday, September 16
12:00 – 1:00 p.m.
Lecture Hall 2, Engineering Science Hall (bldg. 9)
A light lunch will be served at 11:45 a.m.
Professor William Tang
Princeton University, Department of Astrophysical Sciences, Plasma Physics Section, Executive Committee
Princeton Institute for Computational Science & Engineering (PICSciE)
Principal Research Physicist, Princeton Plasma Physics Lab
“Deep Learning Acceleration of Progress toward Delivery of Fusion Energy”
Abstract:
Professor Bill Tang of Princeton University’s Plasma Physics Laboratory will present an innovative application of machine learning on some of the world’s most powerful supercomputers, including the new globally number one ranked “Summit” machine, to detect the onset of and avoid potentially destructive magnetohydrodynamic instabilities in plasma fusion reactors, such as the $25 billion International Thermonuclear Experimental Reactor (ITER) in France. Cross-cutting benefits abound to other systems too complex for real-time predictive simulation from first principles.
Biography
William Tang is Principal Research Physicist at the Princeton Plasma Physics Laboratory, Lecturer with Rank & Title of Professor in the University’s Dept. of Astrophysical Sciences, member of the Executive Board and PI for the Intel Parallel Computing Center (IPCC) at the University’s interdisciplinary “Princeton Institute for Computational Science and Engineering, and Distinguished Visiting Professor at Shanghai Jiao Tong University’s Center for High Performance Computing and NVIDIA Center of Excellence. He is a Fellow of the American Physical Society, was U.S. PI for the G8 Research Council’s “Exascale Computing for Global Scale Issues” Project in Fusion Energy (2011-14), and received the Distinguished Achievement Award from the Chinese Institute of Engineers-USA (2005), the High Performance Computing Innovation Excellence Award from the International Data Corporation (2013), and the NVIDIA 2018 Global Impact Award. Most recently, he is PI of the new project on “Accelerated Deep Learning Discovery in Fusion Energy Science” that has been selected as one of the DOE-ALCF-21 Early Science Projects: https://www.alcf.anl.gov/articles/alcf-selects-data-and-learning-projects-aurora-early-science-program.
For more information about the lecture, please email Professor David E Keyes.