Computational Hypergraph Discovery

Theo Bourdais

My talk on Computational Hypergraph Discovery! This framework addresses the challenge of discovering unknown functional relationships and hypergraph structures from partial observations using Gaussian Processes.

The approach introduces a kernel generalization of Row Echelon Form reduction and variance-based analysis to unveil hypergraph structures, with applications to network discovery in gene pathways, chemical systems, and mechanical systems.

Presentation Venues

DTE AICOMAS
February 18, 2025
Paris, France
SIAM Mathematics of Data Science (SIAM MDS)
October 24, 2024
Atlanta, Georgia, USA
Digital twins for inverse problems in Earth science
July 23, 2024
Marseille, France
SIAM UQ 2024
March 29, 2024
Trieste, Italy
Differential Equations for Data Science 2024 (DEDS2024)
February 19, 2024
Online
One World Seminar Series on the Mathematics of Machine Learning
January 17, 2024
Online