Research

Publications & Preprints

	Claire Brécheteau and Edouard Genetay and Timothee Mathieu and Adrien Saumard Topics in Robust Statistical Learning. 2022, submitted to ESAIM: Proceedings and Surveys	[Preprint]
	Claire Brécheteau and Aurélie Fischer and Clément Levrard Robust Bregman Clustering. Annals of Statistics 2021, Vol. 49, No. 3, 1679-1701	[Paper]
	Claire Brécheteau and Clément Levrard A k-points-based distance for robust geometric inference. Bernoulli 2020, Vol. 26, No. 4, 3017-3050	[Paper]
	Claire Brécheteau Robust anisotropic power-functions-based filtrations for clustering. Symposium on Computational Geometry 2020, 23:1-23:15	[Paper]
	Claire Brécheteau. A statistical test of isomorphism between metric-measure spaces using the distance-to-a-measure signature. Electronic Journal of Statistics 2019, Vol. 13, No. 1, 795-849	[Paper]
	Claire Brécheteau Robust shape inference from a sparse approximation of the Gaussian Trimmed Loglikelihood.	[Preprint]

Popularization

Partitionner des données corrompues, à l'aide de divergences de Bregman
21/04/2022, in Actualités scientifiques, La Lettre de l'INSMI, [Actualité]




Invitation to International Conferences

Workshop on “Optimal Transport, Topological Data Analysis and Applications to Shape and Machine Learning”
Approximating data with unions of ellipsoids for clustering, with intrinsic dimensionality selection.
july 2020, University Columbus, Ohio, USA

SIAM Conference on Applied Algebraic Geometry (AG17)
Distance to a measure for a comparison of poitclouds.
aug 2017, Atlanta, USA

European Meeting of Statisticians (EMS 2019)
Robust shape inference from a sparse approximation of the Gaussian trimmed loglikelihood. jul 2019, Palermo, Italy

PhD Thesis

Towards a Robust Vision of Geometric Inference
Claire Brécheteau
Université Paris-Saclay - 2018
[pdf] [slides]

Some R code content

Here is some code for the experiments in the paper [4] : file

Here is some tutorial (in french) for the experiments in the paper [2] : html version , Gitlab projet with correction_textes_auteurs.Rmd file

Some Python code

Some Python code is available for the robust distance approximations of papers [3] and [6].
The jupyter notebook is available here: file.html and file.ipynb .