Autumn School - Scientific Machine Learning and Dynamical Systems

Lectures and hands-on tutorials for PhD students

9 oktober 2023 van 09:00 tot 13 oktober 2023 16:00 CEST (GMT+0200)
Amsterdam Science Park Congress Centre, Euler room

Scientific Machine Learning and Dynamical Systems

Autumn School jointly organised by NDNS+, 4TU.AMI, CWI

Aim of the event

This event aims at introducing PhD students, postdocs and other early career researchers to the emerging field of Scientific Machine Learning (SciML), in which Machine Learning are used in close connection with physics applications. Numerous leaders in the field will provide lectures on the state-of-the-art in both theory and applications.


Martina Chirilus-Bruckner (Leiden University, the Netherlands)
Nathan Kutz (University of Washington, USA)
Benjamin Sanderse (Centrum Wiskunde & Informatica, the Netherlands)
Mengwu Guo (University of Twente, the Netherlands)

autumn school


The autumn school will revolve around three themes:

  1. Data-driven reduced-order models, such as SINDy, DMD and Koopman theory, as well as non-intrusive and projection-based ROMs.
    Speakers: Steven Brunton, Urban Fasel, Andrea Manzoni, Mengwu Guo, Karen Veroy-Grepl
  2. Neural networks and differential equations, including topics like neural ODEs, differentiable physics, and PDE-inspired neural networks
    Speakers: Max Welling, Erik Bekkers, Hod Lipson, Chris Rackauckas
  3. Data-driven multiscale modeling (coarse graining and closure modeling)
    Speakers: Andrea Beck, Romit Maulik, Wouter Edeling

Each day will consist of an inspirational talk by world experts and a series of "hands-on" lectures that involve both theory and actual coding sessions.

Monday 9 October
Data-driven multiscale methods, turbulence
09:30 - 10:00 Walk-in, coffee
10:00 - 10:10 Welcome and opening
10:15 - 11:15 Andrea Beck: Data-Driven, Discretization Consistent LES models
11:30 - 12:30 Romit Maulik
12:30 - 13:30 Lunch
13:30 - 16:30 Syver Agdestein, Benjamin Sanderse: learning neural closure models for fluid flows
with coffee break at 15:00

Tuesday 10 October
Reduced order models
09:00 - 10:30 Andrea Manzoni, Mengwu Guo
10:30 - 11:00 Coffee break
11:00 - 12:00 Karen Veroy
12:00 - 13:00 Lunch
13:00 - 15:30 Mengwu Guo, Andrea Manzoni
15:30 - 17:30 Poster session and drinks

Wednesday 11 October
SINDy, DMD, Koopman
09:00 -10:00 Steven Brunton
10:00 - 10:30 Coffee break
10:30 - 12:30 Urban Fasel
12:30 - 13:30 Lunch
13:30 - 14:30 Urban Fasel
14:30 - 15:00 Coffee break
15:00 - 17:00 To be decided in discussion with participants

Thursday 12 October
PDE-inspired neural networks
10:00 - 11:00 Paris Perdikaris
11:00 - 11:30 Coffee break
11:30 - 12:30 Erik Bekkers
12:30 - 13:30 Lunch
13:30 - 16:00 Erik Bekkers
with coffee break at 15:30
±17:30 dinner @Maslow

Friday 13 October
Neural ODEs, differentiable physics
09:00 - 12:30 Chris Rackauckas
with coffee break at 10:30
12:30 - 13:30 Lunch
13:30 - 14:00 Overview of approaches and key learning points
14:00 - 15:00 Hod Lipson

Registration for this full-week event is closed. The fee is 75 euros and covers coffee, lunches, and the joint dinner.
As we have a limited number of slots, we might need to do a selection. For this purpose, please give a short motivation why you want to attend the Autumn School. If successful, your attendance will be confirmed soon after July 1st, upon which you will be requested to pay the registration fee.

Note that a background in mathematics, computer science or a related discipline will be advantageous, as well as knowledge in areas such as PDEs, Dynamical Systems, Scientific Computing and Machine Learning.

For students who are short on travel budget, we will try to cover hotel costs for staying in Amsterdam. Please send an e-mail to Nada Mitrovic if you would like to apply for this, including a short motivation.

Interested participants may make a hotel reservation at Hotel Casa (Eerste Ringdijkstraat 4, Amsterdam). Hotel Casa is conveniently located near Amstel train station and from there you can walk, cycle or take the bus to CWI.

Autumn School jointly organised by