Scientific Machine Learning
CIRM, Marseille, France
July 17 - August 25, 2023
Brief presentation of the projects on Monday, July 24
On the first morning of the Hackathon session we will organise a quick presentation (5~10 min) of each project. The purpose of these presentations is to inform all the participants about the other projects and to foster scientific discussions between the groups.Daily seminars
Daily seminars will take place during the Hackathon session, once every morning at 9am before the different groups start working on their own project.The format will be 45min + 15min for questions.
If you wish to share your most recent research in scientific machine learning, do not hesitate to reach out to the organizers at cemracs23@smai.emath.fr.
If you are a project supervisor or sponsor, you are very welcome to share your research or the activites of your company while being at CIRM.
When contacting the organizers, please choose an available date from the calendar at the bottom of the page.
The accommodation and meals at CIRM have to be booked separately (booking details coming soon).
Final presentation of the projects
The last three days will be devoted to the final presentation of the projects.Tentative schedule
Monday | Tuesday | Wednesday | Thursday | Friday | |
---|---|---|---|---|---|
Week 1 |
24/07
Initial presentation of the research projects (9:00 - 12:15) |
25/07 (9:00 - 10:00)
Benedikt Brantner (MPI für Plasmaphysik) Geometric Numerical Integration and Scientific Machine Learning |
26/07 (9:00 - 10:00)
Genevieve Dusson (CNRS, Université Franche Comté) Toward nonlinear reduced basis method using optimal transport in electronic structure calculations |
27/07 (9:00 - 10:00)
Nicolas Trafny (Naval Group) Multidimensional integration using machine learning and Monte Carlo methods for acoustic predictions |
28/07 (9:00 - 10:00)
Thibault Faney (IFPEN) Learning how to accelerate numerical simulations |
Week 2 |
31/07 (9:00 - 10:00)
open |
01/08 (9:00 - 10:00)
Tor Harald Sandve (NORCE) From physics to data-driven reservoir simulations |
02/08 (9:00 - 10:00)
Thibault Malou (INRAE) Localization of pest insects by solving an inverse problem and pheromon detection |
03/08 (9:00 - 10:00)
Stéphane Lanteri (Inria) Bridging high order methods and machine learning for nanoscale photonics |
04/08 (9:00 - 10:00)
open |
Week 3 |
07/08 (9:00 - 10:00)
open |
08/08 (9:00 - 10:00)
Paul Novello (IRT Saint Exupery) Uncertainty analysis as an ally for deep-learning-based hybridization of simulation codes |
09/08 (9:00 - 10:00)
Bruno Després (Sorbonne Université) Algèbre des fonctions ReLU, séries convergentes et algorithmes |
10/08 (9:00 - 10:00)
Victorita Dolean (Université Côte d'Azur) Multilevel domain decomposition-based architectures for physics-informed neural networks |
11/08 (9:00 - 10:00)
Virginie Grandgirard (CEA) TBA |
Week 4 |
14/08 (9:00 - 10:00)
open |
15/08
closed (holiday) |
16/08 (9:00 - 10:00)
Bertrand Iooss (EDF) Methods for global sensitivity analysis of computer codes |
17/08 (9:00 - 10:00)
open |
18/08 (9:00 - 10:00)
Virginie Ehrlacher (ENPC) Sparse approximation of quantum optimal transport problems: application to quantum chemistry |
Week 5 |
21/08 Clément Gauchy (CEA) Surrogate modeling of numerical simulators using Gaussian process regression |
22/08 (15:30 - 16:30)
Olivier Lafitte (Université Paris 13) Une structure logique pour l'apprentissage supervisé: construire un classifieur résultant analytiquement et déduire des propriétés de stabilité |
23/08 final presentation of projects (9:00 - 12:15 and 14:00 - 17:15) |
24/08 final presentation of projects (9:00 - 12:15 and 14:00 - 17:15) |
25/08
open |