UiT Machine Learning Group

Pushing the frontier

Powered by the cool Arctic air, and located at 70° north, the core strength of the Machine Learning Group at UiT The Arctic University of Norway is in basic research for advancing statistical machine learning & AI methodology to face the societal and industrial data-driven challenges of the future.

We publish in CVPR, AAAI, ECCV, IEEE TPAMI, IEEE TNNLS, MICCAI, Medical Image Analysis, to name a few.

We head SFI Visual Intelligence, a Centre for Research-based Innovation funded by the Research Council of Norway and consortium partners.

The UiT Machine Learning Group is hosting the annual Northern Lights Deep Learning Workshop. Please see: nldl.org


  • Robert Jenssen gave an overview of the priorities and activities of Visual Intelligence for health research for the Board of the Northern Norway Regional Health Authority (Helse Nord).

  • Talk by Robert on ML in health 11th March 2021 at a conference organized by the Norwegian Hospital Pharmacists' Association.

  • New paper at CVPR: Reconsidering Representation Alignment for Multi-view Clustering (25% acceptance rate).

  • The new Centre for Research-based Innovation that is hosted at UiT, Visual Intelligence, has officially opened! Please check out visual-intelligence.no for more info.

  • Professor Robert Jenssen guest at the new podcast series from Norwegian Artificial Intelligence Research Consortium (NORA), "NORA forklarer kunstig intelligens". Go to podcast page.

  • Professor Robert Jenssen presented Visual Intelligence's strategy for research and development at the Norway Health Tech's webinar 26. November. View webinar.

  • Michael Kampffmeyer from the UiT Machine Learning Group awarded FRIPRO grant (ground breaking research) from the Research Council of Norway!

  • New paper at AAAI: Measuring Dependence with Matrix-based Entropy Functional (21% acceptance rate).

  • Robert was a panelist at the NORA.startup kick-off: [link to recording].

  • New paper! A novel Dissimilarity Measure for Prototypical Few-Show Learning Networks accepted for ECCV.