publications

2026

  1. L4DC
    Learning to accelerate distributed ADMM using graph neural networks
    Henri Doerks, Paul Häusner, Daniel Hernández Escobar, and Jens Sjölund
    In Learning for Dynamics & Control Conference (L4DC), 2026
    Code
  2. L4DC
    Warm-starting active-set solvers using graph neural networks
    Ella J. Schmidtobreick, Daniel Arnström, Paul Häusner, and Jens Sjölund
    In Learning for Dynamics & Control Conference (L4DC), 2026
    Code

2025

  1. Adv. Theory
    Simul.
    The Frequency-Domain Lattice Boltzmann Method (FreqD-LBM): A Versatile Tool to Predict the QCM Response Induced by Structured Samples
    Diethelm Johannsmann, Paul Häusner, Arne Langhoff, Christian Leppin, Ilya Reviakine, and Viktor Vanoppen
    Advanced Theory and Simulations, Mar 2025
    Code Website
  2. NLDL
    Learning incomplete factorization preconditioners for GMRES
    Paul Häusner, Aleix Nieto Juscafresa, and Jens Sjölund
    Northern Lights Deep Learning Conference, Jan 2025
    arXiv Code

2024

  1. TMLR
    Neural incomplete factorization: learning preconditioners for the conjugate gradient method
    Paul Häusner, Ozan Öktem, and Jens Sjölund
    Transactions on Machine Learning Research, Oct 2024
    arXiv Code Poster

2023

  1. TAG-ML
    ICML 2023 Topological Deep Learning Challenge: Design and Results
    Mathilde Papillon, Mustafa Hajij, Florian Frantzen, Helen Jenne, Johan Mathe, Josef Hoppe, and 60 more authors
    In Proceedings of 2nd Annual Workshop on Topology, Algebra, and Geometry in Machine Learning (TAG-ML), 2023
    PDF Code

2022

  1. MSc Thesis
    Explainable Product Configuration for Telecom via Constraint Reasoning
    Paul Häusner
    Master Thesis, Uppsala University, Feb 2022
    PDF Slides

2019

  1. BSc Thesis
    A comparative evaluation of top-down pruning techniques for Artificial Neural Networks
    Paul Häusner
    Bachelor Thesis, Universität Passau, Apr 2019
    PDF Code