|
Wednesday, June 18, 2025 (location: Auditorium 3, Teorifagbygget hus 6, Google Maps )
|
| 11:00 - 12:00 |
Registration and Light Lunch
|
| 12:00 - 12:15 |
Welcome and Introduction
Robert Jenssen and Kerstin Bach
|
| 12:15 - 13:00 |
Keynote:
Elisabeth Wetzer (UiT The Arctic University of Norway)
Representation Learning for Multimodal Image Registration and Retrieval
Abstract:
Combined information from different imaging modalities enables an integral view of a
specimen, offering complementary information about a diverse variety of its
properties. To efficiently utilize such heterogeneous information, spatial
correspondence between acquired images has to be established. The process is
referred to as image registration and is highly challenging due to complexity, size,
and variety of multimodal biomedical image data.
In this talk, I will give an overview of commonly used techniques from classic image
analysis and learning-based approaches, their limitations and how to efficiently
combine tools from both worlds, particularly when very small training data is
available.
Biography:
Elisabeth Wetzer is associate professor in machine learning in the Machine Learning Group at UiT The Arctic University of Norway and part of the SFI Visual Intelligence and SFF Integreat. Prior to joining UiT in 2024, she did her PhD in Image Processing at Uppsala University in Sweden and worked as a bioinformatician in translational studies on the tumor microenvironment at Karolinska Institute, Sweden. Her research focuses on developing deep learning methodology, with particular focus on biomedical application. As most of these applications suffer from small training data or limited access to labels, she works on methodology which exploits domain context, symmetries in the data, or constraints which may be drawn from other modalities or classic image processing techniques.
Show biography ...
|
| 13:00 - 14:00 |
Technical Session I: Full Papers
Xue-Cheng Tai, Hao Liu, Raymond H. Chan and Lingfeng Li: New Ways to Design Deep Neural Networks
Marius Aasan and Adín Ramírez Rivera: Pixel-Level Predictions with Embedded Lookup Tables
Ratnabali Pal: NorViVQA: Visual Question Answering for Visually Impaired in Norwegian Language
|
| 14:00 - 14:30 |
Coffee Break
|
| 14:30 - 15:30 |
Technical Session II: Full Papers
Yan Zhou, Baifan Zhou and Ingrid Yu: NMN-BART: Generating Natural Language Explanations for Visual Question Answering
Yohanes Nuwara and Markus Hays Nielsen: Role of Prompt Engineering on High-Quality Extraction System from Unstructured Technical Reports in Petroleum Industry using LLM
Tsegazeab Hailu Tedla, Sindre Søpstad and Veralia G. Sanchez: Enhancing the Precision Agriculture of AgriSenze™ by Predicting the Soil Temperature at Different Depths
|
| 15:30 - 15:50 |
Technical Session III: Position Papers
Yuanwei Qu, Arild Waaler, Anita Torabi and Baifan Zhou: Challenges and Approach: AI for Digitalised Carbon Storage Analysis
Teresa Dorszewski, Lenka Tetková, Robert Jenssen, Lars Kai Hansen and Kristoffer Wickstrøm: A Layerwise Analysis of Concept Emergence in ViTs
Sebastian August Berg and Özlem Özgöbek: XAI-Guided Transformer Fine-Tuning for Audio Classification: Automated Explanation-Driven Training with IFI on ViT-Based AudioMAE
Rafael Nozal and Helge Frediksen: Detecting Illegal Fishing with Variational Autoencoders
|
| 16:00 - 16:10 |
Leg Stretch
|
| 16:10 - 16:20 |
Dissertation Award 2024 (chair: Özlem Özgöbek)
The winner of the Dissertation Award 2024 (Dissertation Award 2024) will receive the
certificate.
|
| 16:20 - 16:50 |
Dissertation Award 2024: Presentation
Title and award winner will be announced later.
|
| 20:00 |
Symposium Dinner
|
|
Thursday, June 19, 2025 (Location: Auditorium 2, Teorifagbygget
hus 1, Google Maps )
|
| 9:00-09:50 |
Technical Session IV: Position Papers
Abdenour Benkrid, Omar Zahra and Istvan Szoke: Digital Twin-Enabled Multi-Robot Systems for Safe and Efficient Nuclear Decommissioning
Klaus Johannsen, Xue-Cheng Tai, Junyong You and Gro Fonnes: Learning and predicting fishing activities from AIS data
Changkyu Choi, Arangan Subramaniam, Nils Olav Handegard, Ali Ramezani-Kebrya and Robert Jenssen: Leveraging Foundation Model Adapters to Enable Robust and Semantic Underwater Exploration
Jing Wang, Songhe Feng, Kristoffer Wickstrøm and Michael C. Kampffmeyer: AdaptCMVC: Robust Adaption to Incremental Views in Continual Multi-view Clustering
Xavier Sánchez-Díaz and Ole Jakob Mengshoel: Visualizing Multimodality in Combinatorial Search Landscapes
Tomasz Szczepanski: Challenges when Calculating Reference Limits for Nerve Conduction Studies using Indirect Methods
Ajay Vishwanath and Marija Slavkovik: Machine Ethics or AI Alignment?
|
| 09:50 - 10:30 |
NAIS General Assembly (chair: Odd Erik Gundersen)
|
| 10:30 - 11:30 |
Coffee Break and Poster Session
|
| 11:30 - 12:15 |
Keynote: Steve Marron
(University of North Carolina)
Data Integration Via Analysis of Subspaces (DIVAS)
Abstract:
A major challenge in the age of Big Data is the integration of disparate data types
into a single data analysis. That is tackled here in the context of data blocks
measured on a common set of experimental cases. Joint variation is defined in terms
of modes of variation having identical scores across data blocks. That allows
mathematically rigorous formulation of individual variation within each data block
in terms of individual modes. These are mathematically defined through modes of
variation with common scores. DIVAS improves earlier methods using a novel random
direction approach to statistical inference, and by treating partially shared
blocks. Usefulness is illustrated using mortality, cancer and neuroimaging data
sets.
Biography:
J. S. (Steve) Marron is the Amos Hawley Distinguished Professor of Statistics and Operations Research, Professor of Data Science and Society, Professor of Biostatistics and Adjunct Professor of Computer Science at the University of North Carolina, Chapel Hill. His research lies in many areas of statistics, data science and machine learning, with a special emphasis on gaining simultaneous insights from very diverse data types, including genomics, genetics, imaging and demographics. He enjoys using deep concepts from diverse mathematical areas including geometry and topology in novel data analyses.
Marron’s PhD was earned at UCLA in 1982, under the supervision of Charles J. Stone. Early in his career he worked on the asymptotics of nonparametric curve estimation, with an emphasis on smoothing parameter selection, jointly with many research leaders including Peter Hall and Wolfgang Haerdle. He moved into interdisciplinary research with some foundational work on “code decay” in software engineering, and the invention of fundamental machine learning methods for analyzing gene expression in cancer research.
Marron coined the term “Object Oriented Data Analysis”, which is a framework for addressing the coming trend for data to not only get big, but also to get complex. OODA draws deeply from diverse areas of mathematics including geometry (critical for e.g. shapes as data points) and topology (shown useful for tree / graph data). These ideas have motivated current research on data integration, that has been feeding into continuing cancer collaborations, as well as research on osteo-arthritis, including analysis of human movement data using amplitude-phase decomposition ideas.
In addition to a large amount of reviewing and associate editing, and many NSF Grants, Marron was a founding Associate Director of SAMSI. He also has an unusually strong mentoring record with 60 graduated PhD students.
Show biography ...
|
| 12:15 - 13:00 |
Networking Lunch
|
| 13:15 - 14:00 |
Public Keynote: Keith
Downing (Norwegian University of Science and Technology (NTNU))
Prediction at the core of both natural and artificial intelligence
Abstract:
Prediction is a cognitive advantage like few others, inherently linked to our ability to survive and thrive. Our brains are awash in signals that embody prediction, across a wide range of temporal and spatial scales. In this lecture, I will investigate the origins and anatomy of natural and artificial neural networks as predictive mechanisms. In addition, links to nascent large-language models (LLMs) will be hard to avoid, since their underlying intelligence appears to emerge from performing simple predictive tasks, albeit with complex attention-based representations and massive transformer architectures.
Biography:
Keith Downing is a professor of Artificial Intelligence (AI) at the Norwegian University of Science and Technology (NTNU) in Trondheim, Norway. He is also an
Affiliate Scholar in the Neuroscience Department of Oberlin College. His main interests, experiences and publications are at the intersection of AI, Artificial Life (ALife), and Computational Neuroscience. He spends most of his time either a) teaching several of NTNU’s AI classes, b) programming neural networks and evolutionary algorithms, or c) lecturing to the general public on technical and ethical aspects of AI. Keith's AI and ALife research culminated in two books: "Intelligence Emerging" (MIT Press, 2015) and "Gradient Expectations" (MIT Press, 2023).
Show biography ...
|
Picture: Jakob Bjørvig Henriksen
Picture: Kjetil Rydland
Picture: Jakob Bjørvig Henriksen
