Tuesday, 27 March 2018
4:45pm - 5:30pm
Campus Stuttgart-Vaihingen, Pfaffenwaldring 47
Room V 47.03
Deep Neural Networks in Simulation Science
During the last few years, the rapid development of deep neural networks has led to remarkable advances in image classification, machine translation, and automatic game playing to name a few. However, for the simulation of complex physical/biological systems these advances remain less utilized
In this presentation we discuss ways in which the remarkable power of deep neural networks in areas of inference and classification can be harvested to address problems that challenge traditional simulation techniques. Through a few concrete examples, we illustrate the potential benefits of a collaborative approach, comprising a combination of classic simulations and deep neural networks, and highlight challenges and open questions associated with such an approach.
This work has been done in collaboration with D. Ray (EPFL, CH), S. Ubbliani (USI, CH) and J. Yu (Beihang Uni, China)
After receiving his PhD in 1995 from the Technical University of Denmark, Professor Hesthaven joined Brown University, USA where he became Professor of Applied Mathematics in 2005. In 2013 he joined EPFL as Chair of Computational Mathematics and Simulation Science and since 2017 he serves as Dean of Basic Sciences. His research interests focused on the development, analysis, and application of high-order accurate methods for the solution of complex time-dependent problems, often requiring high-performance computing. Recently he has contributed to the development of reduced basis methods, lately in combination with techniques from machine learning.
He has received several awards for both his research and his teaching, and has published 4 monographs and more than 125 research papers. He is on the editorial board of 8 journals and serves as Editor-in-Chief of SIAM J. Scientific Computing.