Explore a one-hour seminar presentation that delves into the theoretical foundations of neural networks from a numerical analysis perspective. Learn how methods for ordinary differential equations (ODEs) and partial differential equations (PDEs) can provide deeper insights into understanding neural network behavior and properties. Gain valuable knowledge about the mathematical connections between numerical methods and parametric models, bridging the gap between empirical success and theoretical understanding in the field of neural networks. Delivered by Mr Davide Murari from NTNU at the Isaac Newton Institute, the talk addresses the growing need for stronger theoretical frameworks to explain the remarkable practical achievements of neural networks in various applications.