Tabea Rebafka

Assistant professor at LPSM, Sorbonne Université Director of the Master in Statistics at Sorbonne Université

Contact

Email tabea.rebafka AT upmc.fr Office Tour 15-16, 2ème étage, pièce 14 Phone ++ 33 - (0)1 44 27 80 05

Postal address LPSM Sorbonne Université Boîte courrier 158 4 place Jussieu 75252 Paris cedex 05 France

Research Activity

Research Topics

• Random graphs
• Algorithms
• Nonparametric estimation

Research projects

• ANR, EcoNet - Advanced statistical modelling of ecological networks, since 2018
• ANR, BASICS - Bayesian nonparametrics, uncertainty quantification and random structures, since 2017
• PEPS, Spectral methods for machine learning with Malika Kharouf and Nataliya Sokolovska. Project Coordinator. 2017
• PEPS, Random matrix theory for machine learning with Malika Kharouf and Nataliya Sokolovska. Project Coordinator. 2016

Organized conference

• Statistical Advances for Real Data Problems on February 8, 2018 at Jussieu web page

Publications

Journal articles

• Graph inference with clustering and false discovery rate control. With Etienne Roquain and Fanny Villers preprint
• Properties of the Stochastic Approximation EM Algorithm with Mini-batch Sampling. With Estelle Kuhn and Catherine Matias. Statistics and Computing (2020) pdf
• A semiparametric extension of the stochastic block model for longitudinal networks. With Catherine Matias and Fanny Villers. Biometrika, Vol. 105, Iss. 3, p. 665 - 680 (2018). pdf. R code ppsbm-files.zip
• Discussion on `Sparse graphs using exchangeable random measures' by F. Caron and E. B. Fox. With Ismael Castillo. Journal of the Royal Statistical Society, Series B, Vol. 79, No. 5, p. 1295 - 1366 (2017). pdf
• Nonparametric weighted estimators for biased data. With Fabienne Comte. Journal of Statistical Planning and Inference, 174, 104-128 (2016). pdf
• Nonparametric Estimation of the Mixing Density Using Polynomials. With François Roueff. Mathematical Methods of Statistics, Vol. 24, No. 3, 200-224 (2015). pdf
• OMP-type Algorithm with Structured Sparsity Patterns for Multipath Radar Signals. With Céline Lévy-Leduc and Maurice Charbit. Technical report (2011). pdf
• Adaptive Density Estimation in the Pile-up Model Involving Measurement Errors. With Fabienne Comte. Electronic Journal of Statistics, 6, 2002-2037 (2012). pdf
• Information bounds and MCMC parameter estimation for the pile-up model. With François Roueff and Antoine Souloumiac. Journal of Statistical Planning and Inference, 141, 1–16 (2011). pdf
• A Corrected Likelihood Approach for the Pile-Up Model with Application to Fluorescence Lifetime Measurements Using Exponential Mixtures. With François Roueff and Antoine Souloumiac. The International Journal of Biostatistics, Vol. 6, Iss. 1, Article 9 (2010). pdf
• Bootstrap-based tolerance intervals for application to method validation. With Stéphan Clémençon and Max Feinberg. Chemometrics and Intelligent Laboratory Systems, 89, 69–81 (2007). pdf

Patent

• Procédé d’estimation des paramètres de la distribution des temps de réponse de particules d’un système, appliqué notamment aux mesures de fluorescence. With François Roueff and Antoine Souloumiac. Patent Number 09 00524, February 2009.

Proceedings

• Inférence de graphe avec contrôle du taux de faux positifs. With Fanny Villers and Etienne Roquain. In: Journées de Statistique de la SFdS, 2020. pdf
• Eigenrange: A Robust Spectral Method for Dimensionality Reduction. With Malika Kharouf and Nataliya Sokolovska. EUSIPCO, 2018. in press
• Estimation adaptative de densité dans un modèle de transformation nonlinéaire. With Fabienne Comte. In: 43e Journées de Statistique de la SFdS, Gammarth, Tunisie. pdf
• Regularization Methods for Intercepted Radar Signals. With Céline-Lévy Leduc and Maurice Charbit. In: IEEE Radar Conference 2011, Kansas City, USA.
• Désempilement de mesures de temps de réponse par un algorithme E.M. modifié. With François Roueff and Antoine Souloumiac. In: GRETSI 2009: 22ème colloque sur le traitement du signal et des images, Dijon, France. pdf
• An MCMC approach for estimating a fluorescence lifetime with pile-up distortion. In: GRETSI 2007: 21ème colloque sur le traitement du signal et des images, Troyes, France. pdf

Thesis

• Estimation in the Pile-Up Model with Application to Fluorescence Lifetime Measurements.
  Ph.D. thesis defended on 23 october 2009. pdf

Code

• R and Matlab code for the Paper Properties of the Stochastic Approximation EM Algorithm with Mini-batch Sampling. With Estelle Kuhn and Catherine Matias. Statistics and Computing (2020) tgz
• R code for the variational EM-algorithm for the Poisson process stochastic block model with Catherine Matias and Fanny Villers: ppsbm-files.zip This file contains the analysis of three datasets with PPSBM.
• R package ppsbm released on CRAN. With Daphné Giorgi, Catherine Matias and Fanny Villers, 2018. This package contains the optimized R code for PPSBM.

Enseignement

2020-2021

• OIP, Directrice d'étude en M1, SU
• Probability Refresher, Master Data Science for Business, X-HEC

Slides pdf

Textbook pdf

• Probabilités numériques et statistique computationnelle, M1, avec Vincent Lemaire

Poly pdf

Notebook 6, Estimation de paramètres de la loi Gamma, NB6_sujet.ipynb, NB6_sujet.html

Notebook 7, Bootstrap, NB7_sujet.ipynb, NB7_sujet.html

Notebook 8, Le package boot, NB8_sujet.ipynb, NB8_sujet.html

Notebook 9, Modèle de mélange, NB9_sujet.ipynb, NB9_sujet.html

Notebook 10, Algorithme EM pour mélange gaussien, NB10_sujet.ipynb, NB10_sujet.html

Notebook 11, Echantillonneur de Gibbs pour mélange gaussien, NB11_sujet.ipynb, NB11_sujet.html

• Analyse statistique de graphes, M2 de Statistique

Slides Séance 1+2, Séance 3+4, Séance 4-6

TD Feuille 1, Feuille 2

TP Séance 1, Corrigé 1, Séance 2, Corrigé 2, Séance 3

Poly version 2017 pdf

Chapitre 1, version 2018 pdf

• Aléatoire - MAP361, Ecole polytechnique

Feuilles d'exercices PC 1, PC 2, PC 3, PC 4, PC 5, PC 6, PC 7, PC 8, PC 9

Notes d'anciens cours

• Statistical Modeling with R, Master in the Mathematical Sciences, AIMS Sénégal, 2018

Slides Chapter 1, Chapter 1, Chapter 2, Chapter 3, Chapter 4, Chapter 5

Computer Labs Lab 1, Lab 2, Lab 3,

Exercises Series 1, Series 2, Series 3

Assignment zip

• Statistique, 4M015, M1, version 2017, Partie II
• Statistique appliquée, M1, version 2015, Partie I et II

Notebooks d'anciens cours

• Statistique numérique (3M248)

Notebook 1, Rappels et prise en main, NB1_Series.ipynb, NB1_Series.html

Notebook 2, Tableaux de données, NB2_Dataframe.ipynb, NB2_Dataframe.html

Notebook 3, Figures, NB3_Figures.ipynb, NB3_Figures.html

Notebook 4, Statistique descriptive univariée, NB4_stat_descr_uni.ipynb, NB4_stat_descr_uni.html

Notebook 5, Statistique descriptive pour des données bivariées, NB5_stat_descr_bivarie.ipynb, NB5_stat_descr_bivarie.html

Notebook 6, Variables qualitatives et indépendance, NB6_variables_qualitatives.ipynb, NB6_variables_qualitatives.html

Notebook 7, Analyse en composantes principales, NB7_ACP.ipynb, NB7_ACP.html

Notebook 8, Interprétation de l'ACP, NB8_ACP_bis.ipynb, NB8_ACP_bis.html

• Statistique (4M015)

Prise en main de R pdf

Notebook 1, Notebook R, vecteurs et génération de variables aléatoires, Notebook_1.ipynb, Notebook_1.html

Notebook 2, Type de données, Notebook_2.ipynb, Notebook_2.html , Données zip

Notebook 3, Représentations graphiques des données, Notebook_3.ipynb, Notebook_3.html , données anorexie

Programmation en R pdf

Notebook 4, Simulations Monte-Carlo, Notebook_4.ipynb, Notebook_4.html

Notebook 5, Simulations Monte-Carlo (suite), Notebook_5.ipynb, Notebook_5.html

Notebook 6, Régression simple, Notebook_6.ipynb, Notebook_6.html , Données ozone, cathedral

Notebook 7, Régression multiple, Notebook_7.ipynb, Notebook_7.html , Données poidsnaiss

Notebook 8, Analyse de la variance, Notebook_8.ipynb, Notebook_8.html

Planning des salles à Jussieu

Salles de cours et de TD: ici

Salles de TP : ici

Les métiers de la statistique

Zoom sur les métiers de la statistique par la la SFdS: ici