Scientific Machine Learning in Academia and Beyond:
From Theory to Real-World Impact (in Industry)

Still to go
Time until the event
0
0
0
0
Days
0
0
Hrs
0
0
Min
0
0
Sec

Save the date: June 17, 2025, 12.30-17.30

Crowne Plaza hotel, Utrecht 

Scientific machine learning (SciML) is transforming the way we understand, model, and predict
complex systems — and the momentum is only growing. By blending the power of modern
machine learning with the rigor of traditional scientific modelling — including numerical
simulations and physical laws — SciML is opening exciting new frontiers. Faster, smarter, and
more adaptable than ever, these methods are especially impactful in data-scarce domains
where expert knowledge is essential.

Breakthroughs in operator learning, physics-informed neural networks, graph neural networks,
and neural surrogate models are redefining what’s possible. From climate science to
infrastructure, from engineering systems to environmental modelling, SciML is delivering
smarter simulations, deeper insights, and tighter integration between data and domain
knowledge.

But the real test — and the real opportunity — lies in moving beyond the lab. How do we make
these powerful tools robust, scalable, and interpretable enough for real-world deployment?
How can we integrate them into existing workflows in industry, infrastructure, and applied
research?

Join us in Utrecht for an inspiring afternoon that brings together a vibrant community of
researchers, engineers, and innovators from academia, leading applied institutes like Deltares,
and forward-looking industry players. This workshop — Scientific Machine Learning in Practice
— is all about bridging the gap between cutting-edge research and practical application.

What to expect?

Expect engaging talks and case studies ranging from materials discovery and electromagnetic field modelling to accelerating physics simulations and bridging physics with data. Topics will include operator learning, uncertainty quantification, domain-informed AI, and the evolving relationship between AI and scientific discovery — all with a focus on making SciML work in practice. Join us in Utrecht for an afternoon of connection, discussion, and shaping a reliable, explainable, and impactful future for SciML!

 

Keynote speakers

  • Max Welling (CuspAI and UvA, Amsterdam Machine Learning Lab)
  • Stefan Kurz (ETH Zürich, Seminar for Applied Mathematics; also with Robert Bosch
    GmbH, Bosch Center for Artificial Intelligence)
  • Koen Strien (Ignition Computing)
  • Maxim Pisarenco (ASML)
  • Jan Willem van de Meent (UvA, Amsterdam Machine Learning Lab)

Program

12.30-13.25: Lunch & networking

13.25-13.30: Opening

13.30-14.00: Max Welling (CuspAI and UvA AMLab), ‘’Materials Discovery with AI’’

14.00-14.30: Stefan Kurz (ETH Zürich and Robert Bosch GmbH, Bosch Center for Artificial Intelligence), ‘’Can we learn electromagnetic fields?’’

14.30-15.00: Koen Strien (Ignition Computing), ‘’Accelerating linear solves in physics simulations’’

15.00-15.30: Break

15.30-16.00: Maxim Pisarenco (ASML), ‘’AI Research in ASML: bridging physics and data’’

16.00-16.30: Jan Willem van de Meent (UvA AMLab)‘’What can AI do for Science, and what can Science do for AI?’’

16.30-17.15: Discussion

17.15-18.00: Drinks & networking

Registration

Participation is free, but space is limited. Register by emailing bureau@platformwiskunde.nl  with a request to join the CSc–NL June 17 event.
Deadline: May 15. Confirmation will follow shortly after this date.  The confirmation will include a discussion document to prepare for the final session.

Organisation

Computational Science NL (CSc-NL) with 4TU.AMI SRI Bridging Numerical Analysis and Machine Learning, and the national initiative ‘AI and Mathematics’ (AIM).

Information

Questions can be directed to any of the organisers:

Wil Schilders (w.h.a.schilders@tue.nl), Victorita Dolean (v.dolean.maini@tue.nl), Alexander Heinlein (a.heinlein@tudelft.nl), Silke Glas (s.m.glas@utwente.nl)