The PhD defence will be partially digital, in room 1259, Abels Utsikt - Niels Henrik Abels hus 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 audience to ask questions ex auditorio at the end of the defence. If you would like to ask a question, click 'Raise hand' and wait to be unmuted.
-
Join the disputation
The webinar opens for participation just before the disputation starts, participants who join early will be put in a waiting room.-
Submit request to access the thesis (available from October 31st 2:30 pm until November 15th 2:30 pm)
Trial lecture
November 15th, time: 13:00 am, room 1259 (Abels utsikt) and zoom.
- Join the trial lecture
The webinar opens for participation just before the trial lecture starts, participants who join early will be put in a waiting room.
Computability theory is a branch of mathematics and theoretical computer science that seeks to answer questions about existence of algorithms. More specifically, suppose we are given a computational problem that we wish to solve with a computer. This can, for example, be finding the zeros of a real valued function using Newton’s method, or training a neural network to classify images. From the standpoint of computability theory one then wishes to answer the following question: For a given input class, can I compute a solution for each input? We say that a problem is non-computable if the answer to the question above is “no”.
This thesis is a collection of three papers that each study the computability theory of some applied problems. More specifically, two of the papers are concerned with the limitations of artificial intelligence, the computability theory of deep learning methods, and the prevalence of non-computable problems in deep learning. In these papers we specifically focus on the application of deep learning in inverse problems. In the remaining paper, we study the computability theory of semi-definite programs and we focus on the special case of nuclear norm minimization.
Adjudication committee
- Professor Felix Voigtlaender, KU Eichstätt-Ingolstadt
- Professor Elena Celledoni, Norwegian University of Science and Technology
- Professor Erik Bedos, University of Oslo
Supervisors
- Professor Nadia S. Larsen, University of Oslo
- Professor Anders C. Hansen, University of Cambridge
Chair of defence
Head of Department Geir Dahl.
Host of the session
Professor Erik Bedos, University of Oslo