Thomas Bonald (Télécom Paris) a animé un Seminar@SystemX sur le thème « Spectral Methods for Graph Embedding », le 17 septembre 2020. 

Résumé (en anglais) :

Spectral methods are instrumental for the representation of large graphs in vector spaces of low dimension. In this talk, we will review these methods, their physical interpretation, and highlight the impact of some pre- and post-processing steps (regularization, scaling, normalization). We will also show how to apply these techniques using scikit-network, a Python package for the analysis of large graphs.
Keywords: Graph embedding, Laplacian matrix, spectral decomposition, random projection.

Biographie (en anglais) :

Thomas Bonald is a Professor at Télécom Paris, Institut Polytechnique de Paris and the head of the DIG team (Data, Intelligence, Graphs) at LTCI. His current research interests are in the analysis of graphs, time-series and natural language. He is the author of more than 100 scientific papers and 10 patents, and one of the main developers of scikit-network. He received the Blondel Medal in 2013. Find out more.

Replay :