During his PhD studies, Mikael Møller Høgsgaard theoretically researched the accuracy guarantees that Machine Learning and AI models can obtain given a certain amount of data. One direction of Mikael's PhD studies was exploring the guarantees that ensemble learning—combining multiple models to give an answer—has compared to using a single model. The research conducted has shown that in multiple cases, there is a provable theoretical improvement when combining multiple models over using a single model.

The PhD study was completed under the supervision of Professor Kasper Green Larsen in the Algorithms, Data Structures, and Foundations of Machine Learning Group at the Computer Science Department, Faculty of Natural Sciences, Aarhus University.

Time: Thursday, 27. November 2025 at 11:00
Place: Building 5335, room 295, Department of Computer Science, Aarhus University, Helsingforsgade 12, 8200 Aarhus N.
Title of PhD thesis: Guarantees and Insights in Ensemble Learning
Contact information: Mikael Møller Høgsgaard, e-mail: hogsgaard@cs.au.dk, tel.: +45 61549486
Members of the assessment committee:
Professor Shai Ben-David, Department of Computer Science, University of Waterloo, Canada
Professor Amir Yehudayoff, Department of Computer Science, University of Copenhagen, Denmark
Chair: Professor Ira Assent, Department of Computer Science, Aarhus University, Denmark
Main supervisor:
Professor Kasper Green Larsen, Department of Computer Science, Aarhus University, Denmark
Language: The PhD dissertation will be defended in English