Bilde av Mylius-Kroken, Johan
Photo: Petter Bjørklund
Bilde av Mylius-Kroken, Johan
Doctoral Research Fellow Department of Physics and Technology johan.m.kroken@uit.no +4777644202 Tromsø

Johan Mylius-Kroken


Job description

PhD Fellow in Machine Learning.

My research is about how deep learning models behave and what they learn. Primarily this is done using elements from information theory and statistical mathematics. I have a background from computational astrophysics where I used machine learning techniques in an attempt at finding relativistic effect in the large scale structure formation of the Universe, via simulations.


  • Youssef Wally, Johan Mylius-Kroken, Michael Kampffmeyer, Rezvan Ehsani, Vladan Milosevic, Elisabeth Wetzer :
    Flatland and Beyond: Mutual Information Across Geometries
    IEEE International Conference on Computer Vision Workshop (ICCVW) 19. October 2025 DOI / ARKIV
  • Youssef Wally, Johan Mylius-Kroken, Michael Kampffmeyer, Rezvan Ehsani, Vladan Milosevic, Elisabeth Wetzer :
    Hyperbolic Representation Learning for Spatial Biology: Evaluating Cell Type Hierarchies in Breast Cancer Imaging Data
    05. January 2026 DOI / ARKIV
  • Youssef Wally, Johan Mylius-Kroken, Michael Kampffmeyer, Rezvan Ehsani, Vladan Milosevic, Elisabeth Wetzer :
    Hyperbolic Representation Learning for Spatial Omics
    15. November 2025 ARKIV
  • Youssef Wally, Johan Mylius-Kroken, Michael Kampffmeyer, Rezvan Ehsani, Vladan Milosevic, Elisabeth Wetzer :
    Mutual Information Across Geometries
    25. November 2025 ARKIV
  • Youssef Wally, Johan Mylius-Kroken, Michael Kampffmeyer, Rezvan Ehsani, Vladan Milosevic, Elisabeth Wetzer :
    Hyperbolic Representation Learning for Spatial Biology: Capturing Cell Type Hierarchies in Breast Cancer
    07. December 2025 DOI / ARKIV
  • Johan Mylius-Kroken, Kristoffer Wickstrøm, Elisabeth Wetzer, Ali Ramezani-Kebrya, Robert Jenssen :
    Can a Convex Partition caused by a CPWL Neural Network be used for Density Estimation?
    National conference on image processing and machine learnin 2025 ARKIV
  • Youssef Wally, Johan Mylius-Kroken, Michael Kampffmeyer, Rezvan Ehsani, Vladan Milosevic, Elisabeth Wetzer :
    Presentation at PhD school on multi-modal learning
    11. August 2025 ARKIV

  • The 50 latest publications is shown on this page. See all publications in NVA here →