Fast and scalable computation of shape-morphing nonlinear solutions with application to evolutional neural networks

Open Webpage

William Anderson, Mohammad Farazmand

Published: 01 Jan 2024, Last Modified: 16 May 2025J. Comput. Phys. 2024EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract: Highlights•Symbolic RONS uses symbolic computing to develop a scalable computational method for solving the RONS equations.•Collocation RONS introduces a collocation point approach to RONS.•Regularization of the RONS equations speeds up their time integration by addressing the stiffness issue.•Application to evolutional neural network is investigated and demonstrated on a numerical example.