Ta strona wykorzystuje ciasteczka ("cookies") w celu zapewnienia maksymalnej wygody w korzystaniu z naszego serwisu. Czy wyrażasz na to zgodę?

Czytaj więcej

ML Workshop Physics-Informed Neural Networks and Neural Operators

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) and neural operators.

These workshops are organized by the Machine Learning Center of the University of Warsaw and EUROCC2, represented by ICM of the University of Warsaw.

The registration deadline was on 10 (Friday) XI 2023. The registration has already been closed and disabled.

more information

THE REGISTRATION FORM LINK (disabled)

The Workshop schedule

Day 1 (preliminary, non-obligatory)

25.11.2023 (Saturday) [10:00 – 13:00] (Szymon Nowakowski)

  • Introduction to PyTorch
  • Hands-on exercises:
    • Writing and running Python in Colab
    • Using PyTorch to create and train simple neural networks
  • We will provide one 10 minute break

Materials for Day 1 are available in this repository

Solutions for Day 1 are available in this repository

Day 2 (main session)

02.12.2023 (Saturday) [10:00 – 15:00]

  • 10:00-10:45 Short Lecture (Grzegorz Gruszczyński)
    • Introduction to Physics-Informed Neural Networks (PINNs)
    • The physics-informed loss function
  • 10:45-11:00 break
  • 11:00-11:45 Hands-on exercise: Training a PINN to solve a simple ODE (mass-spring-damper) (Grzegorz Gruszczyński)
  • 11:45-12:45 lunch break
  • 12:45-15:00 Hands-on exercise: Training a PINN for parameter identification in PDE (heat transfer) (Marek Bukowicki)

Materials for Day 2 are available in this repository

 

Day 3 (main session)

09.12.2023 (Saturday) [10:00 – 15:30]

  •  10:00-10:45 Short Lecture (Grzegorz Gruszczyński)
    • Introduction to the concept of Neural Operators
    • Architectures: DeepONet, Fourier Neural Operators, ViTO (Vision Transformer Operator)
  • 10:45-11:00 break
  • 11:00-11:45 Hands-on exercise: Fourier Neural Operator – introduction (Szymon Nowakowski)
  • 11:45-12:45 lunch break
  • 12:45-13:30 Hands-on exercise: Understanding a Fourier layer  (Szymon Nowakowski)
  • 13:30-13:45 break
  • 13:45-14:30 Hands-on exercise: Training a Fourier Neural Operator to learn a function-to-function mapping (Szymon Nowakowski)
  • 14:30-14:45 break
  • 14:45-15:30 Hands-on exercise: Fighting for super-resolution (Szymon Nowakowski)

Materials for Day 3 are available in this repository

 

 

Important information: 

  1. The workshop will be conducted in English.
  2. This is an in-person (no streaming) event only.
  3. Place: Faculty of Physics, Ludwika Pasteura 5, 02-093 Warszawa, University of Warsaw
    https://maps.app.goo.gl/WrnrxeCJc3bMZZK48
  4. Room: 0.06 (this is an interactive map of the building)
  5. The registration deadline was on 10 (Friday) XI 2023. The registration has already been closed.
  6. The number of places is limited. The accepted participant list has been emailed on 13 (Monday) XI 2023.