New Face: Lars Berge Olsen
From : Norway
Start date: 17.08.20
Section: 2 - Statistics and Data Science
I finished my Masters at the University of Oslo in the spring of 2020 under the supervision of Associate Professor Riccardo De Bin. My thesis was within the field of Statistical Boosting and it was titled "Likelihood-Based Boosting: Approximate Confidence Bands and Intervals for Generalized Additive Models". We studied a method for deriving approximate uncertainty measures for boosted models with penalized B-splines or penalized stumps as base-learners, and compared the coverage of the approximate uncertainty measures to empirical counterparts obtained from bootstrap.
For my PhD, I will focus on Explainable Artificial Intelligence (XAI), more precisely, my research will focus on how we can use Shapley values (from game theory) to explain machine learning models and individual predictions. This project is a part of BigInsight, and my supervisor is Professor Ingrid Kristine Glad, alongside my co-supervisors Kjersti Aas and Martin Jullum from NR.