Hi everyone, 
I’m quite new to this field and still learning the basics, so please excuse any beginner questions.
I’m really interested in Physics-Informed Neural Networks (PINNs) and would love to understand how to implement them step by step.Could someone kindly guide me or share some resources/tutorials that explain the process in a beginner-friendly way? I’d truly appreciate any help or direction you can offer.
Thanks in advance!
Sorry, I have never heard of a PINN.
It’s not a “beginner” topic that DLAI covers.
I assume you already have some basic knowledge of how neural networks work, the basic feed forward ones which are discussed in deep learning course 1/2/3 from Andrew Ng.
Then PINNs are easy to learn after that, providing you have some scientific/engineering background.
See package DeepXDE for example, which is used by many academics. They have lots of demos for basic equations. You can learn from them.
Hello @KiarashBaharan! Sorry for the late reply.
To understand PINNs (Physics-Informed Neural Networks), first you need to understand traditional Machine Learning as mentioned by @ff100. Then you can easily go to PINN which is nothing but changing a loss function by incorporating PDEs loss (Partial Differential Equations). Ben Moseley did a pretty good job explaining PINN for a newbie, here.
Moreover, consider this course of Prof. Ulisses Braga-Neto. Also, follow Prof. George Karniadakis on LinkedIn; he is a GOAT (Greatest of All Time) for Scientific Machine Learning (SciML).