Call for Papers

Call for Papers

Deadline for submissions: 16.09.2022

Notification of acceptance: Early to mid November


Please join for the 6th Northern Lights Deep Learning Conference (NLDL) on 10-12 January 2023 in Tromsø, Norway, organized by Visual Intelligence and the UiT Machine Learning Group.


We look forward to gathering the deep learning community again in the cool arctic air for a physical conference, after two years online.


In addition, the NLDL winter school, which is a part of the NORA research school http://nora.ai, starts at Jan 9 and ends at Jan 13 and incorporates events during the main conference days. The winter school includes scientific topics, industry event, women in AI event, and transferrable skills. More information coming soon at http://www.nldl.org/winter-school.


We invite submissions presenting new and original research on all aspects of Deep Learning. The topics include but are not limited to the following:

  • Architecture, concepts and optimization

  • Deep learning for structured and unstructured data

  • Graph neural networks

  • Generative models

  • Bayesian Deep Learning

  • Lightweight / frugal Deep Learning

  • Explainability and interpretability of Deep Learning models

  • Computer vision

  • Natural language processing

  • Deep Learning for signals, images, 3D and hyperspectral images

  • Deep Learning applications to biology and medicine

  • Deep Learning application to environment and ecology

  • Deep Learning applications to Physics

  • Deep Learning for industrial applications


As always, we are happy to have top international speakers. This year, for instance

  • Mark Girolami – University of Cambridge/Alan Turing Institute

  • Mihaela van der Shaar – University of Cambridge/Alan Turing Institute

  • Polina Golland – MIT

and more to come.


A tentative program will be available soon. We hope to see many participants for a nice scientific gathering on the “north pole”, including social events, and hopefully some northern lights

How to submit a contribution

We invite participants to submit in PDF format either

  • a 2-page extended abstract (non-archival) or

  • a 6-page full paper (included in proceedings)

in the area of Deep Learning. References can go beyond the page limit.

All submissions will undergo double-blind peer review. It will be up to the authors to ensure the proper anonymization of their paper. Do not include any names or affiliations. Refer to your own past work in the third-person, or if needed insert a blank reference. Accepted contributions will be accepted as contributing talks or poster presentations.

A ZIP archive with a LaTeX template can be downloaded here.

Reference instructions

  • References need to include a DOI (if it exists)

  • When citing arXiv papers, the authors need to check if the paper has been published. If there is a published version of the paper, it should be cited instead of the arXiv version.

Additional information

  • We allow authors to upload a preprint of the submitted papers to arXiv

  • For the camera ready version of accepted papers, we allow the authors to add a 7th page to their paper in order to address requests from the reviewers

  • At least one author must register for the conference and present their work in person. If the covid-situation should change dramatically, then we will inform about any changes in this policy and will inform the registered participants

Follow the following links to submit your submission:

Proceedings

There will be a peer-review process for the NLDL submissions and the 6 page contributions are to be considered as 'full' publications. The proceedings will be published through Septentrio Academic Publishing. The proceedings have been accepted as a Level 1 publication in the Norwegian system.

Copyright for accepted papers

Authors of accepted extended abstracts (non-archival submissions) retain full copyright of their work, and acceptance of such a submission does not preclude publication of the same material in another journal or conference.

Papers aimed for proceedings should follow general dual submission policies and submissions that are identical or substantially similar to papers that are in submission to, have been accepted to, or have been published in other archival conferences, journals, workshops, etc. will be rejected. Authors should clearly state any overlapping published or submitted work at the time of submission. Authors should ensure that they are not violating any other venue dual submission policies. Papers in proceedings will be assigned the CCBY 4.0 license.

Questions & Problems

For questions about the submission process or problems with the submission side, please contact Sigurd Løkse: sigurd.lokse@uit.no.