Date: Tuesday 14 June 2022 – 14:00 (Europe/London)
Speaker: Prof Salvatore Cuomo, Associate Professor of Numerical Analysis, University of Naples Federico II
Abstract
A physics-informed neural network (PINN) is employed to approximate the solution of nonlinear partial differential equation systems. In this talk, we present an approach for solving different configurations for the Gray-Scott, a reaction-diffusion system that involves an irreversible chemical reaction between two reactants. Computational results show that the PINN can successfully provide an approximated solution in a variety of conditions, also reproducing the characteristic Turing patterns in the unstable region of the model's parameter space, through a supervised approach that relies on a finite difference method (FDM).
Joint work M.O.D.A.L. Laboratory
Bibliography
[1] Cuomo, S., Di Cola, V. S., Giampaolo, F., Rozza, G., Raissi, M., & Piccialli, F. (2022). Scientific Machine Learning through Physics-Informed Neural Networks: Where we are and What's next. arXiv preprint arXiv:2201.05624.
[2] Raei, M., Cuomo, S., Colecchia, G., & Severino, G. (2021). Solving 3-D Gray–Scott Systems with Variable Diffusion Coefficients on Surfaces by Closest Point Method with RBF-FD. Mathematics, 9(9), 924.
Biography
Salvatore Cuomo is currently Associate Professor at the Department of Mathematics and Applications at the University of Naples Federico II (UNINA). He gained a BA(Hons) in Mathematics in 1997 and a Ph.D. in Applied Mathematics and Computer Science in 2004 from the University of Naples Federico II, Italy.
In 2018 he became an associate professor of Computer Science and Automatic Systems for information processing. He was a visiting Research Scholar at Case Western Research University, Ohio, USA in 2016 and a Visiting Professor at University of Geosciences, Beijing, China in 2019.
He is Co-founder of the Mathematical Modelling and Data Analysis Laboratory M.O.D.A.L (www.labdma.unina.it/modal) research group which deals with issues of Predictive Modelling and Data Analysis.
You can now watch the seminar on YouTube: https://youtu.be/JOftmwzTg34
