Christophe Giraud

Professor at Paris Saclay University

I contribute to the understanding of some fundamental problems in statistics and machine learning, to the analysis of some popular algorithms and to the design of some new ones. I occasionally collaborate with some biologists.

Duty

I am in charge of the Master program Mathematics and applications of Paris Saclay University.

I am also Adjunct Professor at Cornell University and I serve as Action Editor for JMLR.

PhD Students

Publications

Book

  • Introduction to High-Dimensional Statistics (first Edition 2015, second Edition 2021)
book book2
Share your own solutions to the exercises on the wiki-site http://high-dimensional-statistics.wikidot.com/

Preprints

  • Small Total-Cost Constraints in Contextual Bandits with Knapsacks, with Application to Fairness. (with E. Chzhen, Z. Li, and Gilles Stoltz). To appear in NeurIPS 2023 [arXiv].
  • Parameter-free projected gradient descent. (with E. Chzhen and G. Stoltz) [arXiv].
  • Training Integrable Parameterizations of Deep Neural Networks in the Infinite-Width Limit. (with K. Hajjar and L. Chizat) [arXiv].
  • Pair Matching: When bandits meet stochastic block model. (with Y. Issartel, L. Lehéricy, M. Lerasle) [arXiv].
  • PECOK: a convex optimization approach to variable clustering. (with F. Bunea, M. Royer, N. Verzelen) [arXiv]. Most of the materials have been published in "Model assisted Variable Clustering" AOS 2020
  • A pseudo-RIP for multivariate regression (short unpublished note) [arXiv].

Publications

  • Localization in 1D non-parametric latent space models from pairwise affinities. (with Y. Issartel, N. Verzelen) Electronic Journal of Statistics 2023, Vol. 17, No. 1, 1587-1662. [arXiv].
  • The price of unfairness in linear bandits with biased feedback. (with S. Gaucher, A. Carpentier) NeurIPS 2022 [arXiv] [Slides]
  • A Unified Approach to Fair Online Learning via Blackwell Approachability. (with E. Chzhen, G. Stoltz) NeurIPS 2021 (Spotlight) [arXiv] [Slides].
  • Model assisted variable clustering: Minimax-optimal recovery and algorithms (with Florentina Bunea, Xi Luo, Martin Royer, Nicolas Verzelen) Annals of Statistics 2020, Vol. 48, No. 1, 111-137. [Full version with supplementary material on arXiv]
  • Partial recovery bounds for clustering with the relaxed Kmeans. (with N. Verzelen) Mathematical Statistics and Learning 2019, vol.1, No. 3/4 pp.317-374. [arXiv].
  • Bayesian estimation of species relative abundances and habitat preferences using opportunistic data (with C. Coron, C. Calenge and R. Julliard) Environ Ecol Stat (2018) 25: 71. [Link].
  • Capitalizing on opportunistic data for monitoring relative species abundances. (with C. Calenge, C. Coron and R. Julliard) Biometrics 2016 72(2), 649-58. [arXiv].
  • Aggregation of predictors for non stationary sub-linear processes and online adaptive forecasting of time varying autoregressive processes. (with F. Roueff and A. Sanchez-Perez) Annals of Statistics 2015, Vol. 43, No. 6, 2412-2450. [arXiv].
  • A quantitative framework for investigating risk of deadly collisions between marine wildlife and boats (with Q. Sabatier, J. Martin, et al.) Methods in Ecology and Evolution (2016) Volume 7, Issue 1, pages 42–50.
  • Hybridization within Saccharomyces Genus Results in Homoeostasis and Phenotypic Novelty in Winemaking Conditions. ( with T. da Silva , W. Albertin , C. Dillmann, M. Bely, S. la Guerche, C. Giraud, S. Huet, D. Sicard, I. Masneuf-Pomarede, D. de Vienne, P. Marullo) PloS ONE 10(5) Open Access
  • The spatial distribution of Mustelidae in France. (with C. Calenge, J. Chadoeuf, S. Huet, R. Julliard, P. Monestiez, J. Piffady, D. Pinaud, S. Ruette) PLoS ONE 10(3). Open Access
  • Estimator selection in the Gaussian setting. (with Y. Baraud and S. Huet) Annales de l'IHP (2014), Vol. 50, No. 3, pp. 1092-1119. [Download preprint].
  • Yeast Proteome Variations Reveal Different Adaptive Responses to Grape Must Fermentation. (with M. Blein-Nicolas, W. Albertin, B. Valot, P. Marullo, D. Sicard, S. Huet, A. Bourgais, C. Dillmann, D. de Vienne, M. Zivy) Mol Biol Evol. 2013 Jun;30(6):1368-83. [link]
  • Delimiting synchronous populations from monitoring data. (with R. Julliard and E. Porcher) Environmental and Ecological Statistics Vol. 20 (2013), no 3, pp. 337--352. [pdf] .
  • Including shared peptides for estimating protein abundances: a significant improvement for quantitative proteomics. (with M. Blein-Nicolas, H. Xu, D. De Vienne, S. Huet, M. Zivy) Proteomics. (2012) Sep;12(18), pp. 2797--801 [link]
  • High-dimensional regression with unknown variance (with S. Huet and N. Verzelen) Statist. Sci. Volume 27, Number 4 (2012), 500--518. [pdf file] et [Slides].
  • Discussion of "Latent variable graphical model selection via convex optimization" (with A. Tsybakov) Annals of Statistics (2012), v.40, 1984-1988. [pdf file].
  • Graph selection with GGMselect. (with S. Huet and N. Verzelen) Statistical Applications in Genetics and Molecular Biology. Vol. 11 (2012), no3, 1--50. download preprint and the R package GGMselect.
  • Detecting long distance conditional correlations between anatomical regions using Gaussian Graphical Models (with S. Allassonniere and P. Jolivet) Proceeding of MFCA (2011), Toronto, 111--122 [pdf file].
  • Low rank multivariate regression. Electronic Journal of Statistics, Vol. 5 (2011), 775--799. Article (open access), and Download KF R-code.
  • Gaussian model selection with unknown variance. (avec Y. Baraud et S. Huet) Annals of Statistics, Vol. 37 (2009), no2, 630--672. download article, full paper et Slides
  • Mixing Least-square estimators when the variance is unknown. Bernoulli 14, no.4 (2008) 1089--1107. [Download] / Slides
  • Estimation of Gaussian graph by model selection. Electronic Journal of Statistics, Vol. 2 (2008), 542--563. Article (open access) and Slides
  • Gravitational clustering and additive coalescence. Stoch. Proc. Appl. 115 (2005), no. 8, 1302--1322 (download preprint)
  • On the convex hull of a Brownian excursion with parabolic drift. Stoch. Proc. Appl. 106 (2003), no. 1, 41--62 (download preprint)
  • On a shock front in Burgers turbulence. J. Statist. Phys. 111 (2003), no. 1-2, 387--402 (download preprint)
  • Some properties of Burgers turbulence with white noise initial conditions. Probabilistic Methods in Fluids, 161--178, World Scientific Publisher (2003) (download preprint)
  • On regular points in Burgers turbulence with stable noise initial data. Ann. Inst. H. Poincaré Probab. Statist. 38 (2002), no. 2, 229--251 (download preprint)
  • Clustering in a self-gravitating one-dimensional gas at zero temperature. J. Statist. Phys. 105 (2001), no. 3-4, 585--604 (download preprint)
  • Statistics of a flux in Burgers turbulence with one-sided Brownian initial data. (En collaboration avec J. Bertoin et Y. Isozaki) Comm. Math. Phys. 224 (2001), no. 2, 551--564 (download preprint)
  • Genealogy of shocks in Burgers turbulence with white noise initial velocity. Comm. Math. Phys. 223 (2001), no. 1, 67--86 (download preprint)
  • Turbulence de Burgers et agrégation de particules lorsque l'état initial est aléatoire.: thèse de doctorat (Ph D Thesis)

Software


Lecture Notes

  • Mathematics for Artificial Intelligence. M1 Math-IA Université Paris-Saclay [pdf file]
  • Fondements mathématiques de l'apprentissage statistique. Journées mathématiques X-UPS 2012. Editions de l'Ecole Polytechnique [pdf file]
  • Stochastic Calculus [pdf file]
  • Martingales pour la finance. Livre pédagogique, preprint [pdf file]

Vulgarisation

  • C. Calenge, M. Albaret, F. Léger, J.-M. Vandel, J. Chadoeuf, C. Giraud, S. Huet, R. Julliard, P. Monestiez, J. Piffady, D. Pinaud, and S. Ruette. Premieres cartes d'abondance relative de six mustélidés en france: modélisation des données collectées dans les « carnets de bord petits carnivores » de l'ONCFS. Faune Sauvage, (130): 17, 2016.
  • Mathématiques pour la planète terre. Ouvrage collectif. Nouveau Monde Edition, 2014.

Review

Teaching

Master


Ecole Polytechnique

  • MAP311 Probabilités.
  • MAP433 Statistiques.
  • MAP553 Apprentissage statistique.
  • MAP563 Modèles aléatoires en écologie et évolution.
  • MAP574 Méthodes statistiques pour la biologie et l'écologie.
  • MAP594 Stages d'options "modélisation probabiliste et statistiques".

Master Maths of University of the Philippines Diliman

Vitae

Academic Positions

Since 2012: Professor at Université Paris Saclay (Orsay)
Since 2016: Adjunct professor at Cornell University
2012-2018 : Associate Professor at Ecole Polytechnique
2008-2012 : Professeur chargé de cours at Ecole Polytechnique
2007-2008 : Research fellow at INRA Jouy-en-Josas
2002-2008: Maître de conférences at Université de Nice Sophia-Antipolis


Education

2011: Habilitation à diriger des recherche from Université Paris XI
2001: PhD in Mathematics from Université Paris VI (supervisor: Jean Bertoin)
1996-2000: Student from Ecole Normale Supérieure de Paris

Contact

Office: 3A7
Email: christophe.giraud@universite-paris-saclay.fr
Adress: Institut de Mathématiques d'Orsay, Université Paris-Saclay, F-91405 Orsay Cedex, FRANCE
Phone: (+33) 1 69 15 60 29

Links

Google scholar
Accès IMO
La mécanique des flûtes