Inverse Physics Informed Neural Networks (I-PINNs)
Model Physical Systems Parameters With AI
Ratings 4.84 / 5.00
What You Will Learn!
- Understand the Theory behind PDEs equations solvers.
- Build numerical based PDEs solver.
- Understand the Theory behind Inverse-PINNs PDEs solvers.
- Build an Inverse-PINNs code solver.
Description
This comprehensive course is designed to equip you with the skills to effectively utilize Inverse Physics-Informed Neural Networks (IPINNs). We will delve into the essential concepts of solving partial differential equations (PDEs) and demonstrate how to compute simulation parameters through the application of Inverse Physics Informed Neural Networks using data generated by solving PDEs with the Finite Difference Method (FDM).
In this course, you will learn the following skills:
Understand the Math behind Finite Difference Method.
Write and build Algorithms from scratch to sole the Finite Difference Method.
Understand the Math behind partial differential equations (PDEs).
Write and build Machine Learning Algorithms to solve Inverse-PINNs using Pytorch.
Write and build Machine Learning Algorithms to solve Inverse-PINNs using DeepXDE.
We will cover:
Pytorch Matrix and Tensors Basics.
Finite Difference Method (FDM) Numerical Solution for 1D Burgers Equation.
Physics-Informed Neural Networks (PINNs) Solution for 1D Burgers Equation.
Total variation diminishing (TVD) Method Solution for 1D Burgers Equation.
Inverse-PINNs Solution for 1D Burgers Equation.
Inverse-PINNs for 2D Navier Stokes Equation using DeepXDE.
If you lack prior experience in Machine Learning or Computational Engineering, please dont worry. as This course is comprehensive and course, providing a thorough understanding of Machine Learning and the essential aspects of partial differential equations PDEs and Inverse Physics Informed Neural Networks IPINNs.
Let's enjoy Learning PINNs together
Who Should Attend!
- Engineers and Programmers whom want to Learn Inverse-PINNs