Daniel Johansen Trosten defended his PhD thesis “Improving Representation Learning for Deep Clustering and Few-shot Learning” on August 23rd 2023 at UiT The Arctic University of Norway.
Daniel Johansen Trosten defended his PhD thesis “Improving Representation Learning for Deep Clustering and Few-shot Learning” on August 23rd 2023 at UiT The Arctic University of Norway.
Summary of thesis:
The amounts of data in the world have increased dramatically in recent years, and it is quickly becoming infeasible for humans to label all these data. It is therefore crucial that modern machine learning systems can operate with few or no labels.The introduction of deep learning and deep neural networks has led to impressive advancements in several areas of machine learning. These advancements are largely due to the unprecedented ability of deep neural networks to learn powerful representations from a wide range of complex input signals. This ability is especially important when labeled data is limited, as the absence of a strong supervisory signal forces models to rely more on intrinsic properties of the data and its representations.
This thesis focuses on two key concepts in deep learning with few or no labels. First, we aim to improve representation quality in deep clustering - both for single-view and multi-view data. Current models for deep clustering face challenges related to properly representing semantic similarities, which is crucial for the models to discover meaningful clusterings. This is especially challenging with multi-view data, since the information required for successful clustering might be scattered across many views. Second, we focus on few-shot learning, and how geometrical properties of representations influence few-shot classification performance. We find that a large number of recent methods for few-shot learning embed representations on the hypersphere. Hence, we seek to understand what makes the hypersphere a particularly suitable embedding space for few-shot learning.
Evaluation Committee:
- Professor Klaus-Robert Müller, Technical University of Berlin, Germany (1. Opponent)
- Professor Anne H Schistad Solberg, Institute for Informatics, University of Oslo, Norway (2. Opponent)
- Professor Fred Godtliebsen, Department of Mathematics and Statistics (internal member and leader of the committee)
Supervisors:
- Associate Professor Michael Kampffmeyer, Department of Physics and Technology, UiT (main supervisor)
- Professor Robert Jenssen Department of Physics and Technology, UiT (co-supervisor)
- Researcher Sigurd Løkse, Department of Physics and Technology, UiT(co-supervisor)
The Defense was led by Professor John Sigurd Mjøen Svendsen, Pro-Dean at the Faculty of Science and Technology at UiT.
Visual intelligence og norske forskningssentre innen kunstig intelligens er klare til å bidra til regjeringens ambisiøse milliardsatsning på KI for å utvikle trygg, pålitelig, og effektiv kunstig intelligens som vil gagne samfunnet, innovasjonen og utdanningen.
Maskinlæringsgruppa ved UiT skal i gang med å lage en permanent utstilling om kunstig intelligens. Målet er å rekruttere fremtidens forskere innen faget.
Visual Intelligence researcher Kristoffer Wickstrøm was invited to give lectures at the Technical University of Denmark's summer school on self-supervised learning.
Ledig stilling som kommunikasjonsrådgiver i forskningsruppe for maskinlæring. Vi søker engasjert formidler som kan øke vår synlighet ved å formidle vår forskning til et bredt publikum.
Visual Intelligence’s advances in fundamental and applied deep learning research is gaining widespread attention and visibility.
Visual Intelligence-lead course on self-supervised learning as part of the NORA summer school 2023
The Tromsø AI Meetup 2023 brought together AI enthusiasts at Storgata Camping to connect, learn from notable speakers, and explore the exciting developments in AI, fostering a community passionate about responsible innovation in the field.
Changkyu Choi defended his PhD thesis “Advancing Deep Learning for Marine Environment Monitoring” on June 9th 2023 at UiT The Arctic University of Norway.
Visual Intelligence contributed to the Norwegian Parliament’s AI seminar organized by the Health and Care Services Committee
The University Hospital of North Norway, partner in Visual Intelligence, in the media: "AI is the path forward". Read the full story (in Norwegian).
Fjernstyrte undervannsroboter er et viktig verktøy i kartlegging, utforskning og overvåkning av verdens havområder. I slutten av april fikk 22 elever fra videregående skoler i Troms og Finnmark designe, bygge og anvende fjernstyrte undervannsroboter.
A pitch day was organized at UiT for to connect students with companies for external master’s projects and job opportunities in computer science and machine learning.
Master thesis on automation in the fish processing industry using machine learning by Josten Henriksen.
Prof. Pierre Baldi visited Visual Intelligence in Tromsø, February 16-17, 2023. The two days were filled with fruitful meetings on potential scientific collaboration between UCI and Visual Intelligence.
Machine learning phd students from Visual Intelligence present at UiT The Arctic University of Norway's open day 2023, for recruitment of new students.
VIGS is Visual Intelligence's social and scientific initiative for phd students, postdocs and early career researchers! May 3-5 VIGS will organize a joint trip to UiT's field station in Indre Troms. Sign up and read more about the social and scientific program in out event!
The Northern Lights Deep Learning Conference, Tromsø, January 9th to 13th, was a great gathering of 250 deep learning researchers.
Stine Hansen defended her PhD thesis “Leveraging Supervoxels for Medical Image Volume Segmentation With Limited Supervision” on Dec. 16th at UiT The Arctic University of Norway.
7-8 November VIGS organized a successful gathering for early career researchers in Tromsø. The program was filled with both a social and technical program and gathered around 20 researchers from the research partners in Visual Intelligence.
We had the great pleasure of hosting Lars Kai Hansen, Technical University of Denmark, and Irina Voiculescu, University of Oxford, at the Visual Intelligence centre in Tromsø.