Disputation: Andreas Oslandsbotn

Doctoral candidate Andreas Oslandsbotn at the Department of Informatics, Faculty of Mathematics and Natural Sciences, is defending the thesis Scaling kernel-based learning for big data for the degree of Philosophiae Doctor.

Picture of the candidate

Photo: Private

The PhD defence will be partially digital, in Hans Petter Langtangen Lecture Hall and streamed directly using Zoom. The host of the session will moderate the technicalities while the chair of the defence will moderate the disputation.

 

Ex auditorio questions: the chair of the defence will invite the attending audience in Hans Petter Langtangen Lecture Hall to ask ex auditorio questions. 

Trial lecture

"Continual/Lifelong Learning: Problem Formulation and Main Approaches"

Time and place: January 26, 2024 1:00 PM, Simula Research Laboratory, Hans Petter Langtangen Lecture Hall (Kristian Augusts Gate 23)/ Zoom

 

Main research findings

Kernel methods are popular due to their solid and well-understood theoretical foundation. However, kernel-based learning methods generally have considerable memory and computational requirements that scale poorly with the number of training samples. Consequently, in real-world applications, these methods have largely fallen out of favor in the machine learning community with the rise of alternative methods such as multi-layer neural networks. This thesis focuses on developing algorithms that improve kernel methods' memory and computational requirements. The main strategy is creating kernel methods compatible with modern computational models such as parallelization, distribution, and streaming. We develop a novel multi-resolution learning scheme for streaming data (StreaMRAK) and demonstrate it in a biomedical setting. Furthermore, we develop an efficient kernel method for non-linear dimensionality reduction and improve existing graph metrics, extending them to the large graph regime. We demonstrate the suitability of the algorithms we develop with numerical experiments, theoretical analysis, and comparison to existing methods in the field.

Adjudication committee:

  • Associate Professor Wenjing Liao, School of Mathematics, Georgia Institute of Technology, USA

  • Assistant Professor Raffaello Camoriano, Department of Control and Computer Engineering, Politecnico di Torino, Italy

  • Adjunct Professor Are Magnus Bruaset, Department of Informatics, University of Oslo, Norway

Supervisors:

  • Research Engineer Nickolas Ivan Forsch, Simula Research Laboratory
  • Assistant Professor Alexander Cloninger, University of California, San Diego 
  • Research Fellow, Zeljiko Kereta, University College London
  • Professor Aslak Tveito, Simula Research Laboratory

Chair of defence:

Professor Xing Cai

Contact information at Department: Pernille Adine Nordby 

Published Jan. 12, 2024 8:00 AM - Last modified Jan. 26, 2024 1:30 PM