Biblio Discovery — Demo Discovery Service

5 results for "Dus Mathias"

Showing 1–5 · 0.50 seconds · RSS

  1. Preprint Full text

    Grassmannian Geometry and Global Convergence of Variable Projection for Neural Networks

    Dus, Mathias

    arXiv · 2026 · English · arXiv

    Training deep neural networks and Physics-Informed Neural Networks (PINNs) often leads to ill-conditioned and stiff optimization problems. A key structural feature of these models is that they are linear in the output-layer parameters and nonlinear in the hiddenlayer parameters, yielding a…

  2. Preprint Full text

    Wasserstein Formulation of Reinforcement Learning. An Optimal Transport Perspective on Policy Optimization

    Dus, Mathias

    arXiv · 2026 · English · arXiv

    We present a geometric framework for Reinforcement Learning (RL) that views policies as maps into the Wasserstein space of action probabilities. First, we define a Riemannian structure induced by stationary distributions, proving its existence in a general context. We then define the tangent space…

  3. Journal Article

    Spectral stabilization of linear transport equations with boundary and in-domain couplings

    Dus, Mathias; Ferrante, Francesco; Prieur, Christophe

    Comptes Rendus. Mathématique, Vol. 360 No. G3 · 2022 · French · DOAJ Scopus

    In this work, the problem of stabilization of general systems of linear transport equations with in-domain and boundary couplings is investigated. It is proved that the unstable part of the spectrum is of finite cardinal. Then, using the pole placement theorem, a linear full state feedback…

  4. Preprint Full text

    A review of compactness methods for cross-diffusion systems seen as Wasserstein gradient flows

    Dus, Mathias; Jüngel, Ansgar

    arXiv · 2026 · English · arXiv

    A comprehensive methodology for establishing the existence of gradient flows for cross-diffusion systems with respect to suitable energies is proposed. The approach is based on the construction of piecewise-in-time constant approximations via the Jordan-Kinderlehrer-Otto scheme. Compactness of the…

  5. Journal Article Full text

    Comparison between tensor methods and neural networks in electronic structure calculations***

    Dus Mathias; Dusson Geneviève; Ehrlacher Virginie et al.

    ESAIM: Proceedings and Surveys, Vol. 81 · 2025 · English · DOAJ

    This article compares the tensor method density matrix renormalization group (DMRG) with two neural network based methods -namely FermiNet and PauliNet) for determining the ground state wavefunction of the many-body electronic Schrödinger problem. We provide numerical simulations illustrating the…