Udemy – Inverse Physics Informed Neural Networks (I-PINNs) 2023-10 – Download

Description

Inverse Physics Informed Neural Networks (I-PINNs) course. This comprehensive course is designed to give you the skills to use Physics Based Neural Networks (IPINN) effectively. We will cover the basic concepts of solving partial differential equations (PDEs) and show how to calculate simulation parameters by using inverse physics-based neural networks using data generated by solving PDEs using the finite difference method (FDM). We will teach you the following skills in this course:

Understand the mathematics behind the finite difference method.
Write and build algorithms from scratch up to exclusive algorithms using the finite difference method.
Understand the mathematics behind partial differential equations (PDEs).
Write and build machine learning algorithms to solve reverse pins using Pytorch.
Write and build machine learning algorithms to solve inverted pins using DeepXDE.

We will cover:

Pytorch matrix and the basics of tensors.
Numerical solution of the finite difference method (FDM) for the 1D Berger equation.
A physics-informed neural network (PINN) solution for the 1D Berger equation.
Solution of the Total Variation Reduction (TVD) method for the 1D Berger equation.
Inverse pins solution for the 1D Berger equation.
Inverse pins for the two-dimensional Navier-Stokes equation with DeepXDE.

Don’t worry if you don’t have any previous experience in machine learning or computational engineering. This comprehensive and regularly scheduled course provides a thorough understanding of machine learning and fundamental aspects of partial differential equations (PDE) and IPINN neural networks with inverse physical information.