Accepted posters

The following posters will be presented during the online poster session on Discord.

Poster numberPresenterTitle
1Francesca AzzoliniHeritability curve: A local measure of heritability in family models
2Laura GarrisonInteractive Hierarchical Data Exploration through Dimensional Bundling
3Yilin DuConvolutional neural network & jet images
4Mehdi ElahiEnhanced Movie Recommendation Incorporating Visual Features
5Christiane DuschaProbing atmospheric convection in a local valley system - A data-driven approach combining Eulerian and Lagrangian observations
6Julia RomanowskaDiving into registry data: enhancing epidemiology with bioinformatics and experiments
7Hyeongji KimMeasuring Adversarial Robustness using a Voronoi-Epsilon Adversary
8Pierre GillotScalable Bayesian Network Structure Learning via Maximum Acyclic Subgraph
9Berent LundeInfluence-adjusted gradient boosting
10Francine SchevenhovenImproving weather and climate predictions by supermodeling
11Erlend Raa VågsetRepresenting Homology Geometrically: Algorithms and Complexity
12Hauke BartschA Home for AI Solutions Inside the Hospital - The Research Information System of the Western Norway Regional Health Authority
13Juraj PalenikIso-Contour Tracing for Parameter Space Analysis of Atmospheric Convection Model
15Marta EideCentre for Digital Life Norway (DLN2.0) facilitates transdisciplinary research, innovation, and education in Life Science
16Pooja JoshiArtificial Intelligence in Healthcare: Addressing the challenges in context of economic feasibility
17Yuchong ZhangHuman-centred Machine Learning for Maritime Decision Support Systems

Human-centered Data Science session speakers

We are pleased to confirm the following speakers in the Data-driven Science by Data Science session, on June 2 from 11:00-12:30 CET:

Michael Sedlmair

Bio: Professor in Augmented Reality and Virtual Reality at the Visualization Research Center, University of Stuttgart.

Title: Understanding Data: A Quest for Human-Machine Collaboration

Abstract: Working with and leveraging data has become one of the main pillars of modern science, industry, and society. With the advent of learning-based approaches, there has been a strong focus on how this data can be used to fully automatize certain problems. While full automatization is easy to communicate in the media, the contemporary rhetoric seems to forget that most data problems will continue to necessitate meaningful interfaces and cooperations between humans and machines. Our work focuses on these interfaces, and we seek solutions that help people work with and understand their data by combining approaches from data visualization, human-computer interaction, and computational analysis. In this talk, we will look at different examples of such human-data interfaces, and discuss a model that could help us to characterize the problem space between full automation and human-in-the loop approaches.

Barbara Wasson

Bio: Professor at the Department of Information Science and Media Studies and Director of the Centre for the Science of Learning & Technology (SLATE), University of Bergen

Title: Learning Analytics to Support Learning, Teaching and Educational Practice

Abstract: Learning Analytics (LA) has emerged over the past 11 years as a promising field of research and domain of practice. With roots in AI in Education, Educational Data Mining (EDM), and Big Data, the field comprises research into the challenges of collecting, analysing and reporting data with the specific intent to improve learning and the contexts in which it occurs. Placing the needs of the human in the centre is necessary to impact practice. In this talk I will introduce the field of learning analytics and present results from projects in both the school and higher education sectors.

Christoph Trattner

Bio: Professor at the Department of Information Science and Media Studies and Director of the Research Centre for Responsible Media Technology and Innovation – SFI MediaFutures, University of Bergen

Title: Data Science on the Web for Better Food Decision Making

Abstract: According to the World Health Organization around 80% of cases of heart disease, strokes and type 2 diabetes could be avoided if people were to implement a healthier diet. Computational data analytics approaches have been touted as a valuable asset in achieving the ambitious goal of understanding user behavior and being able to develop intelligent online systems, which can positively influence people’s food choices. In this talk, I will present our research on data science approaches to understand, predict and potentially change food decision making in an online context. First, I will show to what extent online food interactions can be linked to real-world health issues such as obesity on a large-scale. After that, I will show how people upload, bookmark or rate online recipes in large online food communities and how contextual factors and biases such as seasonality, temporality, social context or presentation of recipes have an impact on popularity and how they are perceived. Furthermore, I will reveal to what extent these factors and biases can be exploited to model and predict the users’ online food choices. To conclude, I will present some preliminary work aiming to nudge people towards food choices.

Data-driven Science by Data Science session speakers

We are pleased to confirm the following speakers in the Data-driven Science by Data Science session, on June 2 from 09:00-10:30 CET:

Krishna Agarwal

Bio: Associate Professor in Nanoscopy at the Department of Physics and Technology, UiT The Arctic University of Norway

Title: Artificial intelligence solution for microscopy and nanoscopy

Abstract: Microscopy and nanoscopy present several challenges to adoption of AI. The challenges are often dictated by the lack of ground truth, data deficiency, and the limitations imposed by the physics itself. This talk will present various customized AI paradigms for microscopy and nanoscopy data developed at UiT The Arctic University of Norway.

Renate Grüner

Bio: Director of Research at Helse Bergen and Associate Professor at the Department of Physics and Technology, University of Bergen

Title:

Abstract:

Jan Byška

Bio: Assistant Professor at the Department of Visual Computing, Masaryk University and Adjunct Associate Professor in Visualization at the Department of Informatics, University of Bergen.

Title: Interactive Visual Analysis of Chemical Compound Similarity in Virtual Screening

Abstract: Dimensionality reduction and classification techniques are commonly used to process multidimensional datasets in the modern drug discovery process.
However, the underlying calculations often hinder the interpretability of results and prevent experts from assessing the impact of individual molecular features on the resulting representations.
In this talk, I will describe how multiple coordinated views, linking the low dimensional embeddings with biological activity, selected molecular features, and confidence scores can help to understand complex molecular datasets.

Konrad Tywoniuk

Bio: TMS Group Leader at the Department of Physics and Technology, University of Bergen

Title: Machine learning in particle and nuclear physics

Abstract: Machine learning techniques are currently used ubiquitously in a wide range of modern physics applications from solid-state physics to nuclear and particle physics at high-energy colliders. In this talk, we will give a brief introduction of the principles that should guide ML applications to real-world problems and how such techniques can become efficient tool to guide established techniques, on one hand, and to deal with large and noisy data sets, on the other. We also quote some recent interesting applications of supervised and unsupervised learning in physics.

Machine Learning and AI session speakers

We are pleased to confirm the following speakers in the Machine Learning and AI session, on June 1 from 14:15-15:45 CET:

Samuel Kaski

Bio: Professor in Computer Science at the Department of Computer Science, Aalto University.

Title: Data-analysis with humans

Abstract: Data analysis is usually done by humans or for the use of humans. Why, then, do we not include humans in the models we use for data analysis? I will discuss ways to improve modelling results by taking the human user into account in probabilistic data analysis, by joint modelling of the user and the domain data.

Saket Saurabh

Bio: Professor in Theoretical Computer Science at the Institute for Mathematical Sciences, Chennai and Professor in Algorithms at the Department of Informatics, University of Bergen

Title: Invitation to (Parameterized) Streaming Algorithms

Abstract:

Natacha Galmiche

Bio: PhD candidate at the Department of Informatics, University of Bergen

Title: Revealing multi-modality in weather prediction using machine learning and visualization techniques

Abstract: In weather prediction, stochastic ensemble methods are widely used to provide an overview of physically possible outcomes, where the ensemble spread originates from slightly varied initial conditions as well as model imperfections. The conventional interpretation of the generated ensemble employs mean and standard deviation across ensemble members, which works well for unimodal, Gaussian-like distributions. However, this misleads and discards crucial information in the case of multimodality, i.e., distinct likely outcomes. We present a graph-based approach that reveals multimodality in univariate ensemble-based weather prediction by combining existing clustering methods with a novel concept of life span associated with each cluster.

Sjur Kristoffer Dyrkolbotn

Bio: Professor in Civil Engineering, Western Norway University of Applied Sciences

Title: Causality, relevance and counterfactuals

Abstract: Even for a simple and deterministic system with less than ten interdependent variables, there is no consensus on how to define counterfactual causality. Hence, there is also no (adequate) consensus on how to assess the relevance of different variables to the veracity or otherwise of propositions pertaining to their values. I argue that current approaches to causality in computer science cast a counterfactual net that is too wide, resulting in a notion of relevance that is too permissive in many contexts. By contrast, I suggest that a formalisation of legal causality can produce better results and – hopefully – an easier path to an informed consensus for important model classes where causal judgements are also legally and ethically salient.

Data Science and Statistics session speakers

We are pleased to confirm the following speakers in the Data Science and Statistics session, on June 1 from 16:15-17:45 CET:

Mia Hubert

Bio: Professor in Statistics and Data Science at the Department of Mathematics, KU Leuven

Title: Online teaching statistics for future data scientists

Abstract: For many years I have been teaching a master course on multivariate statistics and regression to students in Mathematics and Engineering. This year I developed a substantial make-over of the course. One reason was to make it better fit within the current online teaching world. But the main motivation was to direct it more towards data science. I will explain and show how I tried to combine theoretical concepts with applications and data visualisation.

John Chandler Johnson

Bio: Associate Professor at BI Norwegian Business School and Chief Scientific Officer at GAINSystems, Inc.

Title: Learning from Students: Executive Education Reveals Data Science Implementation Issues and Solutions

Abstract: Five years ago, we launched an executive education program with the objective to teach mid-career professionals enough data science to successfully collaborate with data science partners. We have trained more than 250 people in over 200 organizations throughout Europe, Asia, and North America. Engaging with experienced individuals in diverse organizations and industries reveals ubiquitous, non-modeling issues that consistently handicap data-driven projects. If domain experts understood these common issues and their implications, we anticipate that doomed projects would be abandoned earlier and ultimately successful projects would be completed more quickly. This paper discusses what we have learned from our students, and offers insights to help domain experts become more productive data science partners.

Ketil Malde

Bio: Researcher at the Institute for Marine Research and Associate Professor in Machine Learning at the Department of Informatics, University of Bergen

Title: The arduous voyage from accurate species recognition to sustainable marine resource management

Abstract: To address large-scale data challenges in marine science, we need to
automate data analysis and processing. There is now a large selection
of machine learning tools and methods that are readily available for
use. However, social and organizational obstacles hinder
effective deployment and thus progress towards the overall goal of
better marine resource management.

Geir Nilsen

Bio: PhD candidate at the Department of Mathematics, University of Bergen

Title: Epistemic Uncertainty Quantification in Deep Learning Classification by the Delta Method

Abstract: We propose a low cost variant of the Delta method applicable to L2-regularized deep neural networks based on the top K eigenpairs of the Fisher information matrix. A demonstration of the methodology is presented using a TensorFlow implementation, and we show that meaningful rankings of images based on predictive uncertainty can be obtained for two LeNet-based neural networks using the MNIST and CIFAR-10 datasets. Further, we observe that false positives have on average a higher predictive epistemic uncertainty than true positives.

Call for posters (closed)

We encourage all participants to submit abstracts for the poster session on the 1st of June (see conference programme). We welcome contributions from all areas of data science and its applications, as well as posters that have been presented previously at other conferences.

Please submit your poster abstracts by Sunday 2nd of May

Update 3 May:

We have reached full capacity for on-site poster presentations. There are still some places available for online-only posters. If you would like to submit an abstract for an online-only poster, then please submit your abstract to cedas2021.posters@uib.no by 5 May 2021.

Update 6 May:

The call for posters is now closed.

Closing keynote by Mattias Villani

We are delighted that Prof Mattias Villani has agreed to give the closing keynote talk at the CEDAS conference 2021.

Bio: Mattias is a Professor of Statistics at the Department of Statistics at Stockholm University and at the Department of Computer and Information Science and the Division of Statistics and Machine Learning at Linköping University.

Title: Bridging the gap between data-centric disciplines – an undergraduate education perspective

Abstract: The increased availability of data and computing power has given birth to a number of new data-centric disciplines, each with a somewhat different focus, methodology and culture. I will contrast statistics and machine learning in an attempt to identify areas where changes in undergraduate education can bring data-centric disciplines closer together.

Opening keynote by Chris Holmes

We are delighted that Prof Chris Holmes has agreed to give the opening keynote talk at the CEDAS conference 2021.

Bio: Chris is a Professor in biostatistics at the Department of Statistics and the Nuffield Department of Medicine at the University of Oxford. He is also playing an instrumental role in what are arguably two of the most exciting data science initiatives world-wide: the Alan Turing Institute, where he is the Programme Director of the Health and medical sciences programme, and Health Data Research UK, where he is the Health Data Science and AI Lead.

Title: Prediction and Model Evaluation