Robert Jenssen
YOU?
Author Swipe
View article: From Colors to Classes: Emergence of Concepts in Vision Transformers
From Colors to Classes: Emergence of Concepts in Vision Transformers Open
Vision Transformers (ViTs) are increasingly utilized in various computer vision tasks due to their powerful representation capabilities. However, it remains understudied how ViTs process information layer by layer. Numerous studies have sh…
View article: Random Window Augmentations for Deep Learning Robustness in CT and Liver Tumor Segmentation
Random Window Augmentations for Deep Learning Robustness in CT and Liver Tumor Segmentation Open
Contrast-enhanced Computed Tomography (CT) is important for diagnosis and treatment planning for various medical conditions. Deep learning (DL) based segmentation models may enable automated medical image analysis for detecting and delinea…
View article: Proceedings of NORA’s annual conference 2025
Proceedings of NORA’s annual conference 2025 Open
NORA - The Norwegian Artificial Intelligence Research Consortium - works to strengthen Norwegian research, education and innovation within the fields of AI, machine learning and robotics. NORA is a collaboration between 8 universities, 5 u…
View article: SuperCM: Improving semi-supervised learning and domain adaptation through differentiable clustering
SuperCM: Improving semi-supervised learning and domain adaptation through differentiable clustering Open
Semi-Supervised Learning (SSL) and Unsupervised Domain Adaptation (UDA) enhance the model performance by exploiting information from labeled and unlabeled data. The clustering assumption has proven advantageous for learning with limited su…
View article: EnLVAM: Enhanced Left Ventricle Linear Measurements Utilizing Anatomical Motion Mode
EnLVAM: Enhanced Left Ventricle Linear Measurements Utilizing Anatomical Motion Mode Open
Linear measurements of the left ventricle (LV) in the Parasternal Long Axis (PLAX) view using B-mode echocardiography are crucial for cardiac assessment. These involve placing 4-6 landmarks along a virtual scanline (SL) perpendicular to th…
View article: Reconsidering Explicit Longitudinal Mammography Alignment for Enhanced Breast Cancer Risk Prediction
Reconsidering Explicit Longitudinal Mammography Alignment for Enhanced Breast Cancer Risk Prediction Open
Regular mammography screening is essential for early breast cancer detection. Deep learning-based risk prediction methods have sparked interest to adjust screening intervals for high-risk groups. While early methods focused only on current…
View article: Aggregation of Dependent Expert Distributions in Multimodal Variational Autoencoders
Aggregation of Dependent Expert Distributions in Multimodal Variational Autoencoders Open
Multimodal learning with variational autoencoders (VAEs) requires estimating joint distributions to evaluate the evidence lower bound (ELBO). Current methods, the product and mixture of experts, aggregate single-modality distributions assu…
View article: REPEAT: Improving Uncertainty Estimation in Representation Learning Explainability
REPEAT: Improving Uncertainty Estimation in Representation Learning Explainability Open
Incorporating uncertainty is crucial to provide trustworthy explanations of deep learning models. Recent works have demonstrated how uncertainty modeling can be particularly important in the unsupervised field of representation learning ex…
View article: From Colors to Classes: Emergence of Concepts in Vision Transformers
From Colors to Classes: Emergence of Concepts in Vision Transformers Open
Vision Transformers (ViTs) are increasingly utilized in various computer vision tasks due to their powerful representation capabilities. However, it remains understudied how ViTs process information layer by layer. Numerous studies have sh…
View article: The Conditional Cauchy-Schwarz Divergence With Applications to Time-Series Data and Sequential Decision Making
The Conditional Cauchy-Schwarz Divergence With Applications to Time-Series Data and Sequential Decision Making Open
The Cauchy-Schwarz (CS) divergence was developed by Príncipe et al. in 2000. In this paper, we extend the classic CS divergence to quantify the closeness between two conditional distributions and show that the developed conditional CS dive…
View article: REPEAT: Improving Uncertainty Estimation in Representation Learning Explainability
REPEAT: Improving Uncertainty Estimation in Representation Learning Explainability Open
Incorporating uncertainty is crucial to provide trustworthy explanations of deep learning models. Recent works have demonstrated how uncertainty modeling can be particularly important in the unsupervised field of representation learning ex…
View article: FLEXtime: Filterbank learning to explain time series
FLEXtime: Filterbank learning to explain time series Open
State-of-the-art methods for explaining predictions from time series involve learning an instance-wise saliency mask for each time step; however, many types of time series are difficult to interpret in the time domain, due to the inherentl…
View article: FreqRISE: Explaining time series using frequency masking
FreqRISE: Explaining time series using frequency masking Open
Time-series data are fundamentally important for many critical domains such as healthcare, finance, and climate, where explainable models are necessary for safe automated decision making. To develop explainable artificial intelligence in t…
View article: Generalized Cauchy-Schwarz Divergence and Its Deep Learning Applications
Generalized Cauchy-Schwarz Divergence and Its Deep Learning Applications Open
Divergence measures play a central role and become increasingly essential in deep learning, yet efficient measures for multiple (more than two) distributions are rarely explored. This becomes particularly crucial in areas where the simulta…
View article: Cauchy-Schwarz Divergence Information Bottleneck for Regression
Cauchy-Schwarz Divergence Information Bottleneck for Regression Open
The information bottleneck (IB) approach is popular to improve the generalization, robustness and explainability of deep neural networks. Essentially, it aims to find a minimum sufficient representation $\mathbf{t}$ by striking a trade-off…
View article: Deep-learning-derived input function in dynamic [18F]FDG PET imaging of mice
Deep-learning-derived input function in dynamic [18F]FDG PET imaging of mice Open
Dynamic positron emission tomography and kinetic modeling play a critical role in tracer development research using small animals. Kinetic modeling from dynamic PET imaging requires accurate knowledge of an input function, ideally determin…
View article: LSNetv2: Improving weakly supervised power line detection with bipartite matching
LSNetv2: Improving weakly supervised power line detection with bipartite matching Open
This paper addresses the crucial task of power line detection and localization in electrical infrastructure inspection using Unmanned Aerial Vehicles (UAVs) from weak supervision, polyline annotations. We first identify several limitations…
View article: DIB-X: Formulating Explainability Principles for a Self-Explainable Model Through Information Theoretic Learning
DIB-X: Formulating Explainability Principles for a Self-Explainable Model Through Information Theoretic Learning Open
The recent development of self-explainable deep learning approaches has focused on integrating well-defined explainability principles into learning process, with the goal of achieving these principles through optimization. In this work, we…
View article: A Contextually Supported Abnormality Detector for Maritime Trajectories
A Contextually Supported Abnormality Detector for Maritime Trajectories Open
The analysis of maritime traffic patterns for safety and security purposes is increasing in importance and, hence, Vessel Traffic Service operators need efficient and contextualized tools for the detection of abnormal maritime behavior. Cu…
View article: View it Like a Radiologist: Shifted Windows for Deep Learning Augmentation Of CT Images
View it Like a Radiologist: Shifted Windows for Deep Learning Augmentation Of CT Images Open
Deep learning has the potential to revolutionize medical practice by automating and performing important tasks like detecting and delineating the size and locations of cancers in medical images. However, most deep learning models rely on a…
View article: ADNet++: A few-shot learning framework for multi-class medical image volume segmentation with uncertainty-guided feature refinement
ADNet++: A few-shot learning framework for multi-class medical image volume segmentation with uncertainty-guided feature refinement Open
A major barrier to applying deep segmentation models in the medical domain is their typical data-hungry nature, requiring experts to collect and label large amounts of data for training. As a reaction, prototypical few-shot segmentation (F…
View article: Analysis of Deep Convolutional Neural Networks Using Tensor Kernels and Matrix-Based Entropy
Analysis of Deep Convolutional Neural Networks Using Tensor Kernels and Matrix-Based Entropy Open
Analyzing deep neural networks (DNNs) via information plane (IP) theory has gained tremendous attention recently to gain insight into, among others, DNNs’ generalization ability. However, it is by no means obvious how to estimate the mutua…
View article: Hubs and Hyperspheres: Reducing Hubness and Improving Transductive Few-Shot Learning with Hyperspherical Embeddings
Hubs and Hyperspheres: Reducing Hubness and Improving Transductive Few-Shot Learning with Hyperspherical Embeddings Open
Distance-based classification is frequently used in transductive few-shot learning (FSL). However, due to the high-dimensionality of image representations, FSL classifiers are prone to suffer from the hubness problem, where a few points (h…
View article: On the Effects of Self-supervision and Contrastive Alignment in Deep Multi-view Clustering
On the Effects of Self-supervision and Contrastive Alignment in Deep Multi-view Clustering Open
Self-supervised learning is a central component in recent approaches to deep multi-view clustering (MVC). However, we find large variations in the development of self-supervision-based methods for deep MVC, potentially slowing the progress…