An introduction to physics-informed neural networks (PINNs)
In May, Center4ML (Faculty of Physics) and ICM UW jointly organized a workshop within the EuroCC2 project, titled “ML Workshop: Physics-Informed Neural Networks and Neural Operators”. Recordings from this event are available on our YT channel.
PINNs are a type of machine learning model that can be used to solve partial differential equations (PDEs), which are ubiquitous in physics, engineering, and other fields. PINNs work by embedding the physical laws that govern a given system into the learning process. This allows PINNs to learn accurate solutions to PDEs, even with very spare data.
A detailed event program and materials are available at:
https://indico.icm.edu.pl/event/49/
Learn more:
National Competence Centre in HPC (EuroCC2)
Machine Learning Center of the University of Warsaw (Center4ML)
Day 1 – 11.05.2024
ML Workshop Physics-Informed Neural Networks and Neural Operators [Part 1]
Introduction to PyTorch
Hands-on exercises:
- Writing and running Python in Colab
- Using PyTorch to create and train simple neural networks
Day 2 -18.05.2024
ML Workshop Physics-Informed Neural Networks and Neural Operators [Part 2]
Short Lecture
- Introduction to Physics-Informed Neural Networks (PINNs)
- The physics-informed loss function
Hands-on exercise: Training a PINN to solve a simple ODE (mass-spring-damper)
Hands-on exercise: Training a PINN for parameter identification in PDE (heat transfer)
