Bayesian Contrastive Learning on Topological Structures

Title

Bayesian Contrastive Learning on Topological Structures

 Author

Möllers, Alex (TU Delft Electrical Engineering, Mathematics and Computer Science)

Contributor

Isufi, E. (mentor) ORCID 0000-0002-1919-260X
Kekkonen, H.N. (mentor) ORCID 0000-0003-4700-1724
Fortuin, Vincent (mentor)
Immer, Alexander (mentor)
van Iersel, L.J.J. (graduation committee) ORCID 0000-0001-7142-4706

 Degree granting institution

Delft University of Technology

 Programme

Applied Mathematics | Stochastics

 Date

2023-11-08

 Abstract

In this thesis we develop a Bayesian approach to graph contrastive learning and propose a new uncertainty measure based on the disagreement in likelihood due to different positive samples. Moreover, we extend contrastive learning to simplicial complexes and show that it can be used to generate high-quality representations of edge flow data.

 Subject

Contrastive Learning
Bayesian Contrastive Learning
Simplicial Complexes

 To reference this document use:

http://resolver.tudelft.nl/uuid:0a7fdcf0-2521-4a89-9687-8d1311d5e93d

 Part of collection

Student theses

 Document type

master thesis

 Rights

© 2023 Alex Möllers
Files
PDF master_thesis_amoellers_t ... sitory.pdf 6.77 MB
Close viewer