Adaptive activation functions accelerate convergence in deep and physics-informed neural networks

Ameya Jagtap discusses the nonlinear Klein-Gordon equation, which has smooth solutions, the nonlinear Burgers equation, which can admit high gradient solutions, and the Helmholtz equation.

Image courtesy of Ameya Jagtap

Image courtesy of interviewee

Leave a Reply

Your email address will not be published.

×
Guest visitors to the Faculti website can view any insight monthly. You have insight(s) remaining for this month

Register for free as an individual and view thousands of insights in our archive today or subscribe for subject access.

You may already have institutional access to Faculti. Please access Faculti using your institution credentials.

Copyright © Faculti Media Limited 2022. All rights reserved.
error:

Add the Faculti Web App to your Mobile or Desktop homescreen

Install
×