ML Workshop: Physics-Informed Neural Networks



Grzegorz Gruszczyński (University of Warsaw), Marek Bukowicki (University of Warsaw), Szymon Nowakowski (University of Warsaw)

This workshop is an introduction to physics-informed neural networks (PINNs). 

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.

The Workshops on Machine Learning techniques is for beginners students, PhD candidates, academic staff from fields such as physics, applied mathematics, chemistry, or biology who have a basic understanding of Python and machine learning. During the workshops, the focus will be on Physics Informed Neural Networks (PINNs).

Center4ML: University of Warsaw

EUROCC2: represented by Interdisciplinary Centre for Mathematical and Computational Modelling, University of Warsaw

The workshop will cover the following topics:

Introduction to the concept of PINNs

Mathematical foundations of PINNs

Applications of PINNs in science and engineering

Workshop language:
The workshop will be conducted in English.

Dates (Saturdays): 11 V (10:00-13:30), 18 V (9:00-13:00)

This is an online event. The link to the workshop will be provided by e-mail shortly before the event.

Learning Objectives:
By the end of this workshop, you will be able to:

Understand the basics of Physics-Informed Neural Networks (PINNs) and Neural Operators.

Train PINNs to solve engineering problems.

Train Neural Operators to solve a variety of machine-learning tasks.

The registration will be opened on Monday, April 22, 2024, and closed on May 8 at 23:59.

Szkolenia ICM
    • 10:00 AM 11:00 AM
      Introduction to PyTorch (Szymon Nowakowski)
    • 11:00 AM 11:15 AM
      Break 15m
    • 11:15 AM 1:30 PM
      Hands-on exercises: (Szymon Nowakowski)

      – Writing and running Python in Colab,
      – Using PyTorch to create and train simple neural networks.

    • 9:00 AM 9:45 AM
      Short Lecture (Marek Bukowicki)

      Introduction to Physics-Informed Neural Networks (PINNs),
      The physics-informed loss function.

    • 9:45 AM 10:00 AM
      Break 15m
    • 10:00 AM 10:45 AM
      Hands-on exercise: Training a PINN to solve a simple ODE (mass-spring-damper) (Marek Bukowicki)
    • 10:45 AM 11:00 AM
      Break 15m
    • 11:00 AM 1:00 PM
      Hands-on exercise: Training a PINN for parameter identification in PDE (heat transfer) (Grzegorz Gruszczyński)