ACMD Seminar: Optimization Based Deep Neural Networks with Applications

Harbir Antil
George Mason University

Tuesday, November 9, 2021, 3:00 PM EST (1:00 PM MST)

A video of this talk is available to NIST staff in the Math channel on NISTube, which is accessible from the NIST internal home page.

Abstract: This talk will introduce novel deep neural networks (DNNs) as constrained optimization problems. In particular, the talk will introduce DNNs with memory which helps to overcome the vanishing gradient challenge. The talk will also explore reducing the computational complexity of DNNs by introducing a bias ordering. Approximation properties of the DNNs will also be discussed. 

These proposed DNNs will be shown to be excellent surrogates to parameterized (nonlinear) partial differential equations (PDEs), Bayesian inverse problems, data assimilation problems, with multiple advantages over the traditional approaches. The DNNs will also be applied to chemically reacting flows problems. They require solving a system of stiff ODEs and fluid flow equations. These are highly challenging problems, for instance, for combustion the number of reactions can be significant (over 100) and due to the large CPU requirements of chemical reactions (over 99%) a large number of flow and combustion problems are presently beyond the capabilities of even the largest supercomputers.

Bio: Harbir Antil is the Director of the Center for Mathematics and Artificial Intelligence (CMAI) and a full professor in the Department of Mathematical Sciences at George Mason University. He is the co-Editor-in-Chief of the Springer journal “Advances in Continuous and Discrete Models.” He is on the editorial board of prestigious journals such as “SIAM Reviews”. Antil has also held a research fellowship position at Brown University and is currently an Affiliate Professor at the University of Delaware. His areas of interest include algorithmic optimization, machine (deep) learning, numerical analysis, partial differential equations, and scientific computing with applications in optimal control, shape optimization, dimensional reduction, imaging, fluid dynamics, etc. His research is funded by the National Science Foundation, Airforce Office of Scientific Research (AFOSR), and Department of Navy, and Department of Energy.

Note: This talk will be recorded to provide access to NIST staff and associates who could not be present to the time of the seminar. The recording will be made available in the Math channel on NISTube, which is accessible only on the NIST internal network. This recording could be released to the public through a Freedom of Information Act (FOIA) request. Do not discuss or visually present any sensitive (CUI/PII/BII) material. Ensure that no inappropriate material or any minors are contained within the background of any recording. (To facilitate this, we request that cameras of attendees are muted except when asking questions.)

Host: Günay Doğan