Scientific Machine Learning, emerging topics

Jun 18, 2024, 1:30 PM → Jun 21, 2024, 1:00 PM Europe/Rome

128-129 (SISSA, International School for Advanced Studies, main campus, Trieste, Italy)

"Scientific Machine Learning, emerging topics" is an international conference focused on the study of mathematical theory and algorithms of machine learning, and applications of machine learning in scientific computing and engineering disciplines. Particular emphasis will be on 

• optimization,
• physics-informed learning,
• graph neural networks,
• neural operators, 
• transformers, 
• neural odes, 
• generative models.

 

Conference fees

	Early Bird Fee	Normal Fee
Students	250€	300€
Non students	350€	400€

The conference fee includes: lunches (Wed. and Thu.), coffee breaks, aperitivo, conference dinner and transportation from Trieste City Center to SISSA Campus.

Early Bird Fee until May, the 31st, 2024.

Registration is closed. 

Payment instructions are available at this link. 
SISSA staff will then send the invoices for the payment.

Abstract submission is now closed!

A connection to SISSA campus will be guaranteed by shuttle bus with the following from Piazza Oberdan and SISSA campus:

June 18: 1:00pm (Oberdan) 6:30pm (SISSA)
June 19: 8:45am (Oberdan) 7:30pm (SISSA)
June 20: 8:45am (Oberdan) 6:30pm (SISSA)
June 21: 8:45am (Oberdan) 1:10pm (SISSA)

Moreover, SISSA campus can be reached by local public transportation: the bus 38 connects the city center to the campus every 20 minutes, but passengers need a ticket.

Talks
Keynote talks are supposed to be 1h long (45' talk + 15' questions), other talks are supposed to last 30 minutes (25' talk + 5' questions). 

Poster session details
The poster session will be on Wednesday from 17:00 until 19:00. The poster holder are portrait B0 format. During the poster session, an aperitivo will be served.  

Conference dinner
The conference dinner will take place at Savoy Restaurant, Riva del Mandracchio, 4, 34124 Trieste, TS on Thursday at 20:00.

 

Image taken from the paper by Michael M. Bronstein, Joan Bruna, Taco Cohen and Petar Veličković "Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gauges", arXiv preprint arXiv:2104.13478

 

Sponsors:

     

  
 

BookOfAbstracts.pdf

Tuesday, June 18

1 Registration

2 Opening

3 Deciphering Thrombosis: Advancing Multi-Scale Modeling with THRONE

Speaker: Marco Laudato (KTH Royal Institute of Technology)

4 SOGA: Second Order Gaussian Approximation of Probabilistic Programs

Speaker: Francesca Randone (University of Trieste)

5 Operator Learning for Ionic Models in Biomathematics

Speaker: Edoardo Centofanti (Università degli Studi di Pavia)

4:30 PM Coffee break

6 DeepONet-based inverse approximator in matrix-free applications

Speaker: Caterina Millevoi (University of Padova)

7 Turbulence Subgrid Closure for the Lattice Boltzmann Method via Artificial Neural Networks

Speaker: Giulio Ortali (Eindhoven University of Technology, SISSA International School for Advanced Studies)

Wednesday, June 19

8 Domain Decomposition and Machine Learning - Two Mutually Beneficial Areas

Speaker: Axel Klawonn (Universitaet zu Koeln)

10:30 AM Coffee Break

9 A Reduced Order Approach for Artificial Neural Networks applied to Object Recognition

Speaker: Laura Meneghetti (SISSA)

10 Deep learning approaches for meshless surrogate models in variable geometries

Speaker: Stefano Pagani (MOX-Dipartimento di Matematica, Politecnico di Milano)

11 Deep unfolding for matrix factorizations

Speaker: Erik Chinellato (Università degli Studi di Padova)

12:30 PM Lunch break

12 Delta-PINNs: physics-informed neural networks on complex geometries

Speaker: Simone Pezzuto (Università di Trento)

13 Graph-based machine learning approaches for model order reduction

Speaker: Federico Pichi (SISSA)

14 Learning stability on graphs

Speaker: Antonioreneè Barletta (Spici srl)

4:30 PM Coffee Break

15 Poster Session

Accepted posters:

Andrea Bonfanti
Matteo Tomasetto
Pavan Pranjivan Mehta
Taniya Kapoor
Jonas Actor
Guglielmo Padula
Johannes Müller
Kateryna Morozovska
Federica Bragone
Jonas Nießen
Francesco Giacomarra
Sajad Salavatidezfouli
Kabir Bakhshaei
Luca Pellegrini
Ivan Cucchi

Thursday, June 20

16 Rate estimates of ONets

Speaker: Christoph Schwab (SAM ETH)

10:30 AM Coffee Break

17 Latent Dynamics Networks (LDNets): learning the intrinsic dynamics of spatio-temporal processes

Speaker: Francesco Regazzoni (MOX, Dipartimento di Matematica, Politecnico di Milano)

18 Mesh-informed reduced order models for aneurysm risk prediction

Speaker: Giuseppe Alessio D'inverno (SISSA)

19 Parameterization learning for scattered data adaptive spline fitting

Speaker: Sofia Imperatore (University of Florence)

12:30 PM Lunch Break

20 Solving the inverse problem in complex systems via machine learning: challenges and perspectives

Speaker: Constantinos Siettos (University of Naples Federico II)

21 PINA: a PyTorch Framework for Solving Differential Equations with Deep Learning

Speaker: Dario Coscia (SISSA)

22 Pre-trained physics-informed deep learning-based reduced order models for nonlinear parametrized PDEs in small data regimes

Speaker: Simone Brivio (Politecnico di Milano)

4:30 PM Coffee Break

23 Panel Session

8:00 PM Social Dinner

Friday, June 21

24 Data-driven Inference of Chemical Reaction Networks via Graph-based Variational Autoencoders

Speaker: Francesca Cairoli (Università di Trieste)

10:30 AM Coffee break

25 Discovery of Dirichlet-to-Neumann Maps on Graphs via Gaussian Processes

Speaker: Adrienne Propp (Stanford University)

26 Unfolding the Kalman Filter iterations in a machine learning framework

Speaker: Fabio Marcuzzi (Università degli Studi di Padova)

27 A comparative review of regression techniques for predicting the physical characteristics of ship engines

Speaker: Fatemeh Mohammadizadehsorouei (SISSA)

12:30 PM Closing