List of confirmed participants:

Talks:

• Benedikt Brantner (Max-Planck-Institut für Plasmaphysik, Garching, Germany):

  Crossroads between Geometric Numerical Integration and Machine Learning

• Elie Bretin (INSA Lyon, France):

  Learning mean curvature flows with neural networks

• Emmanuel de Bézenac (ETH, Zürich, Switzerland):

  An operator preconditioning perspective on training in physics-informed machine learning

• Paola Cinnella (Sorbonne Université, Paris, France):

  Bayesian machine learning and multi-model ensembles for data-driven turbulent flow prediction

• Julie Digne (CNRS & LIRIS, Lyon, France):

  Implicit neural representations, geometry and simulation

• Régis Duvigneau (Inria, Sophia-Antipolis, France):

  Physics-Informed Neural Networks for fluid dynamics : opportunities, current issues and possible paths forward

• Virginie Ehrlacher (CERMICS, ENPC, Marne la Vallée, France):

  Numerical solution of eigenvalue Schrödinger problems in high dimension using infinite-width two-layer networks

• Jean Feydy (Inria, Paris, France):

  Computational optimal transport: recent speed-ups and applications

• Borjan Geshkovski (Inria, Paris, France):

  Approximate transport with normalizing flows

• Virginie Grandgirard (CEA Cadarache, IRFM, France):

  Scientific machine learning to optimize plasma turbulence simulations

• Alexander Heinlein (TU Delft, the Netherlands):

  Domain decomposition for neural networks

• Mariella Kast (EPFL, Lausanne, Switzerland):

  Evolutional neural networks for time-dependent parametrised partial differential equations

• Jens Keim (Universität Stuttgart, Germany)

  Reinforcement Learning Based Shock Capturing for Hyperbolic PDEs

• Axel Klawonn (Universität zu Köln, Germany):

  Learning the adaptive coarse basis functions of FETI-DP

• Sebastian Krumscheid (Karlsruher Institut für Technologie, Germany)
• Matthieu Nastorg (Inria Paris, France):

  Multi-Level GNN Preconditioner for Solving Large Scale Problems

• Stefano Pagani (Politecnico di Milano, Italy):

  Implicit neural representations: from static shapes to spatio-temporal dynamical modeling

• Federico Pichi (SISSA, Trieste, Italy):

  Graph-based machine learning approaches for model order reduction

• Gianluigi Rozza (SISSA, Trieste, Italy):

  Accelerating Numerical Simulations by Model Reduction with Scientific and Physics-Informed Machine Learning

• Alena Shilova (Inria, Lille, France):

  Learning HJB Viscosity Solutions with PINNs for Continuous-Time Reinforcement Learning

• Martin Šípka (Univerzita Karlova, Praha, Czech Republic):

  Balancing Constraints and Accuracy in Learning of Poisson Systems

• Wassim Tenachi (Observatoire Astronomique de Strasbourg, France):

  (Machine)-learning analytical physical laws

• Janine Weber (Universität zu Köln, Germany)

  A Domain Decomposition-Based CNN-DNN Architecture for Model Parallel Training of Image Recognition Problems

• Tizian Wenzel (Universität Hamburg, Germany):

  Two-layered and deep kernels and applications in machine learning

Posters:

• Amaury Bélières–Frendo (Université de Strasbourg, France):

  Geometric shape optimization for the Dirichlet energy with physics-informed and symplectic neural networks

• Sofiane Ezzehi (CERMICS, ENPC, Marne la Vallée, France):

  Galerkin Neural Networks: Approximating Variational Equations and Inequalities

• Frédérique Lecourtier (Inria, Strasbourg, France):

  Combining finite element methods and neural networks to solve elliptic problems on complex 2D geometries

• Lucas Mager (Universität zu Köln, Germany):

  A data-driven surrogate model for an efficient homogenization of open-porous materials

• Youssouf Nasseri (Inria, Strasbourg, France):

  Neural network based-closure for fluid model of plasma physics

• Khoa Nguyen (ENS Paris-Saclay, France):

  Development and assessment of PINNs frameworks for multi-geometry simulation in industrial contexts

• Moreno Pintore (Inria, Paris, France):

  Certified upper bounds for the Lipshitz constant of neural networks

• Alexandre Pugin (Inria, Sophia-Antipolis, France):

  PINNs for Maxwell's equations in the time domain

• Guillaume de Romèmont (ONERA, Châtillon, France & ENSAM, Paris, France):

  A data-driven learned discretization approach in finite volume schemes for compressible fluid dynamics

• Claire Schnoebelen (Université de Strasbourg, France):

  Learning Hamiltonian functions of ODEs and PDEs

• Janne Siipola (Helsingin yliopisto, Finland):

  Better sampling with gradient

• Léopold Trémant (Inria, Strasbourg, France):

  Learning non-canonical Hamiltonian dynamics

• Yanfei Xiang (Inria, Bordeaux, France):

  Neural operator preconditioning for accelerating the solution of the parametric Helmholtz equations

Participants:

• Juliette Chabassier (Inria Bordeaux & Modartt, France)
• Raphaël Côte (Université de Strasbourg, France)
• Stéphane Cotin (Inria Strasbourg, France)
• Clémentine Courtès (Inria & Université de Strasbourg, France)
• Ruiyang Dai (Inria, Paris, France)
• Bruno Després (Sorbonne Université, Paris, France)
• Michel Duprez (Inria, Strasbourg, France)
• Victorita Dolean (TU Eindhoven, the Netherlands)
• Emmanuel Franck (Inria, Strasbourg, France)
• Rodrigo Ibata (Observatoire Astronomique de Strasbourg, France)
• Samuel Kokh (SGLS, CEA Paris-Saclay, France)
• Stéphane Lanteri (Inria, Sophia-Antipolis, France)
• Vanessa Lleras (Université de Montpellier, France)
• Victor Michel-Dansac (Inria, Strasbourg, France)
• Hadrien Montanelli (Inria, Paris, France)
• Laurent Navoret (Inria & Université de Strasbourg, France)
• Pierre Ocvirk (Observatoire Astronomique de Strasbourg, France)
• Christophe Prud'homme (Universté de Strasbourg & CEMOSIS, France)
• Nathan Ricard (Inria, Sophia-Antipolis, France)
• Alexis Thibault (Modartt, France)
• Julie Tryoen (Thales, Vélizy-Villacoublay, France)