News archive for Center for Data Science
What I currently think about machine learning in health, is that it will lead to better health in the world I will try to touch several issues, using examples from the literature and mine (this will be mainly breast cancer models for personalized therapy), among which bias, intelligence (the dynamic one and the hard-coded one), explain black box models (or stop using them?) and slow-AI.
HyperTraPS: Learning evolutionary and disease progression pathways from large-scale biological and medical datasets (10.02.2020)
Iain George Johnston will talk about HyperTraPS (hypercubic transition path sampling), a highly generalisable Bayesian approach which efficiently learns evolutionary and progression pathways from cross-sectional, longitudinal, or phylogenetically linked data.
The speaker will focus on the following topics: inferring topological invariants of latent structures from data, enriching dimension reduction layouts, and modeling human perception to design more trustworthy Machine Learning techniques.
With the advances in image editing and image synthetizing techniques, a new phrase well describes the reality: “Seeing is disbelieving”. Duc Tien Dang Nguyen and his team are working to restore truth in visual media.
In this I2S seminar series lecture, Ambra Demontis will talk about how to attack machine learning systems as well as how to defense against those attacks.
Supporting behavioral change using psychologically-aware recommender systems: studies on household energy conservation – and more (10.10.2019)
In this talk, Alain Starke presents his research on energy recommender systems.
The focus of this seminar will be on trajectory inference methods, a novel class of unsupervised computational techniques to model dynamic processes.
Ensemble-based kernel learning for a class of data assimilation problems with imperfect forward simulators (04.09.2019)
In this talk, the speaker will introduce an approach to dealing with simulator imperfection from a point of view of functional approximation that can be implemented through a certain machine learning method, such as kernel-based learning.
Computational science presentation
The talk will give an overview of computational methods for protein structure prediction.
A seminar about how distant reading and network analysis can be used to develop a big picture of cultural discourse about machine vision technologies.
The objective of the workshop is to build bridges between parameterized complexity theory and practical applications.
Data science related seminar on machine learning and visual analytics (speaker: Chaoran Fan).
Norwegian research and education related to artificial intelligence, machine learning and robotics received a boost when Norwegian institutions formed the consortium NORA. The collaboration also stimulates several start-up companies within a field in rapid development.
This meeting is being organized to facilitate collaboration between UiB researchers and NORA (Norwegian Artificial Intelligence Research Consortium) in the fields of artificial intelligence and machine learning.
This meeting is focused on kernelization, the rigorous theory of preprocessing.
On Friday the 5th of April at 10am Jürgen Bernard will give a talk at the Visual Computing Forum. The talk will take place at Lille Aud. @ Høyteknologisenteret (Thormøhlens gate 55). The title of the talk is: Enhancing Human-Centered Machine Learning with Visual Analytics.
Data science related talk by PD Dr. Steffen Oeltze-Jafra