CITru organizes computer science, mathematics and statistics, and law faculty and students from three Norwegian units; UiT in Tromsø, UiB in Bergen, and SimulaMet in Oslo. The distinct profile of CITru is architecting of complex, trustworthy, large-scale digital eco-systems, a continuously hard problem and moving target of highest importance for any modern society.
We particularly focus on how to combine intelligence and trust in next-generation analysis technologies that harness advances in Big Data, Artificial Intelligence (AI), and distributed computing. Digital trust can be at conflict with, for instance, existing opaque approaches to machine learning (ML). Supervised learning, unsupervised learning, and reinforcement learning are different types of ML with each best suited to certain use cases. The typical “black box” approach of such techniques creates a transparency challenge by not showing which factors led to a result and how. Trust in the algorithm depens on transparancy and explainability, and is particularly important in, for instance, criminal justice or financial lending AI applications. How to provide trust in next-generation analytics systems therefore attracts much of the attention in CITru.