Evolving-Graph Gaussian Processes

Thumbnail Image
Access rights
openAccess
Journal Title
Journal ISSN
Volume Title
A4 Artikkeli konferenssijulkaisussa
This publication is imported from Aalto University research portal.
View publication in the Research portal
View/Open full text file from the Research portal
Other link related to publication
Author
Date
2021-07
Department
Department of Electrical Engineering and Automation
Department of Computer Science
Major/Subject
Mcode
Degree programme
Language
en
Pages
6
Series
Abstract
Graph Gaussian Processes (GGPs) provide a dataefficient solution on graph structured domains. Existing approaches have focused on static structures, whereas many real graph data represent a dynamic structure, limiting the applications of GGPs. To overcome this we propose evolvingGraph Gaussian Processes (e-GGPs). The proposed method is capable of learning the transition function of graph vertices over time with a neighbourhood kernel to model the connectivity and interaction changes between vertices. We assess the performance of our method on time-series regression problems where graphs evolve over time. We demonstrate the benefits of e-GGPs over static graph Gaussian Process approaches.
Description
Keywords
gaussian process, Time series, Graph-based learning, probabilistic models
Other note
Citation
Blanco-Mulero, D, Heinonen, M & Kyrki, V 2021, ' Evolving-Graph Gaussian Processes ', Paper presented at International Conference on Machine Learning: Time Series Workshop, Virtual, Online, 24/07/2021 - 24/07/2021 . < https://arxiv.org/abs/2106.15127 >
Permanent link to this item
https://urn.fi/URN:NBN:fi:aalto-202108258360
Collections
[article-cris] Perustieteiden korkeakoulu / SCI