Etienne Lasalle

Research

My research focuses on statistics related to graph-structured data.
During my thesis, I developed and studied tools for multi-scale graph comparisons based on heat diffusion and topological data analysis. These tools come with statistical guarantees that ensure the asymptotic validity of two-sample tests. Implementing these methods allowed me to apply them to practical problems, particularly in machine learning and neural network classifiers.
Currently, I am exploring various aspects of graph data compression. This includes work on graph inference through a compressive learning method and, more recently, investigating methods to accelerate community detection algorithms using coarsening techniques.

Articles

A multilevel approach to accelerate the training of Transformers, 2025.
With G. Lauga, M. Chaumette, E. Desainte-Maréville and A. Lebeurrier
Accepted at GRETSI 2025.
[HAL]
A note on the relations between mixture models, maximum-likelihood and entropic optimal transport, 2025.
With T. Vayer
[arxiv]
PASCO (PArallel Structured COarsening): an overlay to speed up graph clustering algorithms, 2024.
With R. Vaudaine, T. Vayer, P. Borgnat, R. Gribonval, P. Gonçalves, M. Karsai
Accepted at ECML 2025 (Journal Track) and published in Machine Learning.
[Journal] [arxiv]
Compressive Recovery of Sparse Precision Matrices, 2023.
With T. Vayer, R. Gribonval, P. Gonçalves
[arxiv]
Eve, Adam and the Preferential Attachment Tree, 2023.
With A. Contat, N. Curien, P. Lacroix and V. Rivoirard
Published in Probability Theory and Related Fields.
[journal] [arxiv]
Heat diffusion distance processes: a statistically founded method to analyze graph data sets, 2023.
Published in J Appl. and Comput. Topology (SI : Data Science on Graphs).
[journal] [arxiv]

Talks

Material

Others

Editorial activities

Awards