Skip to main content
high contrast version
pl polski menu
"

Breadcrumb Breadcrumb

Web Content Display Web Content Display

Most significant publications

  • Marcin Mazur, Łukasz Pustelnik, Szymon Knop, Patryk Pagacz, Przemysław Spurek, Target layer regularization for continual learning using Cramer-Wold distance, Information Sciences 609 (2022), 1369-1380
  • Szymon Knop, Marcin Mazur, Igor Podolak, Przemysław Spurek, Jacek Tabor, Generative models with kernel distance in data space, Neurocomputing 487 (2022), 119-129
  • Stanisław Jastrzębski, Szymon Knop, Marcin Mazur, Igor Podolak, Przemysław Spurek, Jacek Tabor, Cramer-Wold Auto-Encoder, Journal of Machine Learning Research 21 (2020), 1-28
  • Marcin Mazur, Piotr Oprocha, S-limit shadowing is C^0-dense, Journal of Mathematical Analysis and Applications 408 (2013), 465-475
  • Marcin Mazur, Jacek Tabor, Computational hyperbolicity, Discrete and Continuous Dynamical Systems 29 (2011), 1175-1189

Recent publications

  • Dariusz Jajeśniak, Piotr Kościelniak, Przemysław Klocek, Marcin Mazur, Interpoint Inception Distance: Gaussian-Free Evaluation of Deep Generative Models, International Conference on Computational Science [ICCS](MAIN), (2024), accepted
  • Przemysław Spurek, Sebastian Winczowski, Zięba Maciej, Tomasz Trzcinski, Kacper Kania, Marcin Mazur, Modeling 3D Surfaces with a Locally Conditioned Atlas, International Conference on Computational Science [ICCS](MAIN), (2024), accepted
  • Magdalena Proszewska, Marcin Mazur, Tomasz Trzcinski, Przemysław Spurek, HyperCube: Implicit Field Representations of Voxelized 3D Models (Student Abstract), National Conference of the American Association for Artificial Intelligence [AAAI] 38 (2024), 23623-23625
  • Łukasz Struski, Marcin Mazur, Paweł Batorski, Przemysław Spurek, Jacek Tabor, Bounding Evidence and Estimating Log-Likelihood in VAE, International Conference on Artificial Intelligence and Statistics [AISTATS](MAIN) 206 (2023), 5036-5051
  • Przemysław Spurek, Artur Kasymov, Marcin Mazur, Diana Janik, Sławomir K. Tadeja, Łukasz Struski, Jacek Tabor, Tomasz Trzciński, HyperPocket: generative point cloud completion, IEEE/RSJ International Conference on Intelligent Robots and Systems [IROS], (2022), 6848-6853

Interests

Machine and statistical learning methods, including (but not limited to) generative models, variational inference, and continual learning

avatar for Marcin Mazur

Marcin Mazur

academic degree/title Doctor of Science position
research and faculty staff member group, assistant professor
unit
  • Department of Applied Mathematics
  • Institute of Mathematics
research group
  • Group of Machine Learning Research (GMUM)
ORCID
0000-0002-3440-8173
contact
marcin.mazur@uj.edu.pl