← All plenary speakers

Yanlai Chen

Professor of Mathematics, University of Massachusetts Dartmouth

High-order RBM-inspired linear and nonlinear model order reduction approaches for Physics-Informed Machine Learning

Abstract

Physics-Informed Neural Network (PINN) has proven itself a powerful tool for solving nonlinear partial differential equations (PDEs) leveraging the expressivity of deep neural networks and the computing power of modern heterogeneous hardware. However, its training is still time-consuming, especially in the multi-query and real-time simulation settings, and its parameterization often overly excessive.

In this talk, we present the Generative Pre-Trained PINN (GPT-PINN) and its nonlinear and sparsified variants. Inspired by the traditional Reduced Basis Method, GPT-PINN mitigates both challenges PINNs face in the setting of parametric PDEs in the linear reduction regime. As a network of networks and meta-learning paradigm for parametric systems, its outer-/meta-network is light-weight having significantly reduced number of neurons. Its activation function at each hidden neuron is a (full) PINN pre-trained at a judiciously selected system configuration. After a highly efficient post-training, the meta-network can generate surrogate solutions for the parametric system across the entire parameter domain accurately and efficiently. GPT-PINN's nonlinear and sparsified versions extend the methodology to the nonlinear reduction regime and further speedup the training and inference via orders of magnitude of sparsification done to the collocation set.

Bio

Dr. Chen's work focuses on numerical analysis and scientific computing, with specialized expertise in finite element methods, reduced basis methods, uncertainty quantification, the design of machine learning algorithms and their linear / nonlinear model order reduction. His research is consistently supported by federal agencies, including the National Science Foundation (NSF), the Air Force Office of Scientific Research (AFOSR), and the Office of Naval Research (ONR).

A passionate advocate for education-led economic development and student success, Dr. Chen led the establishment of the NSF-funded ACCOMPLISH program at UMass Dartmouth, which provides contextualized, computing-centered education and comprehensive support for eligible STEM students. Extending his leadership to statewide and regional impact, he also serves on the Board of Directors for the Massachusetts Clean Energy Center (MassCEC) and on the steering committee for the U.S. Economic Development Agency (EDA) - designated Ocean Tech Hub.

Dr. Chen earned his B.S. in Mathematics from the University of Science and Technology of China. He subsequently earned an M.S. in Computer Science and a Ph.D. in Mathematics from the University of Minnesota before completing his postdoctoral fellowship at Brown University. Beyond his academic endeavors, he is a Boston Marathon finisher who enjoys the outdoors and spending time with his family and friends.