Physics-Informed Neural Networks with MATLAB
Temo Vekua and Armando Garcia
MATLAB Staff
March 23, 2026
Email office@phys.ksu.edu for the Zoom address
Abstract
Physics-Informed Neural Networks (PINNs) are an emerging approach that combines deep learning with fundamental physical laws. By constraining neural networks with our domain knowledge, PINNs can produce solutions that are more consistent with known physics and often more accurate and reliable than purely data-driven models.
The session will cover the core concepts behind PINNs and demonstrate how to implement them in MATLAB.
Finally, low-code approaches and GenAI coding assistance are explored to speed up the development process.
Who Should Attend
Faculty, staff, researchers, and students are welcome! Optional free introductory MATLAB Onramp Training here.
About the Presenters
Temo Vekua is the Physics Community Liaison at MathWorks with experience in physics research and lecturing at university level. Before joining MathWorks he was at Indiana University Bloomington, researching dualities in quantum many-body systems.
Armando Garcia is a Customer Success Engineer at MathWorks. In this role, he partners with academics and students to optimize their use of MATLAB for teaching and research. He has an MS in Engineering Mechanics and almost 15 years of combined experience in research and the industry. In the past, Armando has used MATLAB in applications ranging from extracting ultrasonic properties of human skull bone to predicting the performance of pumping equipment.