Program

The meeting is organized around 16 invited talks in the different subjects of expertise of SIGMA. These talks will be accessible to a large audience, and given by international experts. About half of them will discuss new connections with learning theory and data sciences. Furthermore, the meeting will also include contributed talks as well as poster sessions, which will in particular encourage young researchers to present their results.

If you want to attend see the registration/submission page.

Schedule

50min talks are marked with a *, other talks are 30min long. These durations include Q&A

Keynote talks (50min Q&A included)

• Matthieu Dolbeault (RWTH Aachen, Germany),
  Sampling strategies for weighted least-squares (Blackboard talk - no slides)
• Virginie Ehrlacher (École des Ponts, France),
  New low-complexity dynamical approximations for the Schrödinger équation
• Philipp Petersen (University of Vienna, Austria)
• Gabriel Peyré (ENS Ulm, France),
  Transformers are Universal In-context Learners
• Audrey Repetti (Heriot-Watt University, UK),
  Learning approaches for solving monotone inclusion problems in imaging
• Justin Solomon (MIT, USA),
  Navigating, Restructuring and Reshaping Learned Latent Spaces
• Hendrik Speleers (Università degli Studi di Roma, Italy),
  Maximally smooth splines on triangulations
• Claudia Totzeck (University of Wuppertal, Germany),
  Consensus-based optimization - a particle-based alternative for the robust calculation of eigenfaces
• Mario Ullrich (Johannes Kepler University, Austria),
  Widths of convex sets and the power of adaption and randomization
• Adrien Weihs (UCLA, USA),
  Asymptotic Consistency Analysis of (Hyper)graph Algorithms in Semi-Supervised Learning

Invited and contributed talks (30min Q&A included)

• Laurent Baratchart (INRIA Sophia Antipolis, France),
  Least squares rational interpolation to holomorphic functions.
• Raphaël Barboni (ENS / Paris Sciences et Lettre, France),
  Training infinitely deep and wide ResNets with Conditional Optimal Transport
• Simon Boisserée (RWTH Aachen University, Germany),
  Analysis and numerics of nonlinear PDE systems in porous media flow models
• Kristian Bredies (University of Graz, Austria),
  Bregman-based approaches for parameter learning in variational imaging
• Jonathan Chirinos Rodriguez (Université de Toulouse, France),
  Learning firmly nonexpansive operators
• Matthew J. Colbrook (University of Cambridge, UK),
  Towards a classification theory in data-driven dynamical systems
• Julie Delon (Université Paris Cité, France),
  Barycenters for transport costs
• Geneviève Dusson (Université Bourgogne France-Comté, France),
  Copula methods for modeling pair densities in electronic structure calculations
• Salah Eddargani (University of Rome, Italy),
  Quadrature rules for splines of high smoothness on uniformly refined triangles
• Jean Feydy (Inria Paris, HeKA team, France),
  The geometric software stack: past, present, future.
• Jan Groselj (University of Ljubljana, Slovenia),
  Higher-degree super-smooth macro-elements in B-spline-like representations
• Michael Hecht (CASUS / University Wroclaw, Germany),
  The Fast Newton Transform: Interpolation in Downward Closed Spaces
• Didier Henrion (Université de Toulouse, France),
  Solving moment and polynomial optimization problems on Sobolev spaces
• Pierre Marion (EPFL, Switzeland),
  Three stories on deep linear networks
• Gerlind Plonka (University of Goettingen, Germany),
  MOCCA: A Fast Algorithm for Parallel MRI Reconstruction Using Model Based Coil Calibration
• Clarice Poon (University of Warwick, UK),
  Sampling with overparameterized Langevin dynamics
• Camille Pouchol (MAP5 / Université Paris-Cité, France),
  Random Leja points for interpolation
• Nelly Pustelnik (ENS Lyon, France),
  Proximal multilevel algorithm
• Lucia Romani (University of Bologna, Italy),
  A link between Chebyshev polynomials and level-dependent subdivision schemes
• Maria Lucia Sampoli (University of Siena, Italy),
  Hierarchical matrices for 3D Helmholtz problems in multi-patch IgA-BEM setting
• Josua Sassen (ENS Paris-Saclay, France),
  Riemannian Calculus on Latent Manifolds
• Tomas Sauer (Universität Passau, Germany),
  Prony, ideals and Gauss quadrature
• Silvia Sellan (MIT, USA),
  A new perspective on reconstruction from Signed Distance Functions
• Espen Sande (EPFL, Switzerland),
  Optimal spline spaces and outlier-removal strategies in isogeometric analysis
• Christoph von Tycowicz (Zuse Institute Berlin, Germany),
  Manifold GCN: Diffusion-based Convolutional Neural Network for Manifold-valued Graphs
• Tino Ullrich (TU Chemnitz, Germany),
  Nonlinear sampling recovery for multivariate function classes

Posters

• El Mehdi Achour(RWTH Aachen University, Germany),
  A general approximation lower bound in $L_p$ norm, with applications to feed-forward neural networks
• Clémentine Chazal (ENSAE, France),
  Statistical and geometrical properties of the regularized Kernel Kullback Leibler divergence
• Radu Dragomir (Télécom Paris, France),
  Implicit bias of mirror flow for classification on separable data
• Olivier Gibaru (ENSAM, LISPEN, France)
• Clément Guillot (Ecole des Ponts Paristech, France),
  Least square formulation for the time-dependent Schrödinger equation
• Guillaume Houry (INRIA, France),
  Fast 1-Wasserstein Distance Approximations Using Greedy Strategies for Language Processing Applications
• Varun Khurana (UC San Diego, Brown University, USA),
  Neural Network Two-Sample Testing: Detecting Distribution Differences
• Alexandre Leclerc (INSA Rouen, France)
  POKAIOK : A Standalone AI-Powered Assistant for Enhanced Visual Inspection in Production Lines
• Wael Mattar (Tel Aviv University, Israel)
  Pseudo-Reversing and Its Application to Manifold-Valued Multiscaling
• Jiri Minarcik (independent researcher),
  Differentiable Geometric Constraint Enforcement using Minkowski Penalties
• Mimoun Mohamed (Aix Marseille Université, France),
  Straight-Through Meets Sparse Recovery: the Support Exploration Algorithm
• Christophe Rabut (INSA Toulouse, France),
  "de Casteljau algorithm" and "Bézier curves": When? Who? How?
• Uaday Singh (IIT Roorkee, India),
  Some density results by deep Kantorovich type neural network operators
• Viktor Stein (Technical University Berlin, Germany),
  Wasserstein Gradient Flows of Moreau Envelopes of f-Divergences in Reproducing Kernel Hilbert Spaces
• Huqing Yang (RWTH Aachen University, Germany),
  Robust Sparse Low-Rank Approximation of Multi-Parametric PDEs
• Qi Ye (South China Normal University, China),
  Composite Algorithms of Data-driven and Model-driven Methods in Banach Spaces
• Gentian Zavalani (Helmholtz-Zentrum Dresden-Rossendorf, Germany),
  Stable High-Order Approximation of Triangulated Manifolds for Surface Integration