SUPR
Graph Structure Learning for Time-Series
Dnr:

NAISS 2024/22-1641

Type:

NAISS Small Compute

Principal Investigator:

Canberk Ozen

Affiliation:

Högskolan i Halmstad

Start Date:

2024-12-13

End Date:

2026-01-01

Primary Classification:

10201: Computer Sciences

Webpage:

Allocation

Abstract

This project is about learning latent graph structures for time-series datasets, which will then be used for forecasting and time-series anomaly detection tasks with graph neural networks. To achieve this goal, I need to run several experiments trying out existing methods and modifying them for better performance.