Habilité à Diriger des Recherches

Name Yohann De Castro
Mail Institut de Mathématique d'Orsay,
Univ. Paris-Sud, CNRS

Équipe MoKaPlan,
Phone +33 (0)1 69 15 77 97
Office Room 3G1 Building 307
Email yohann.decastro (at) math.u-psud.fr
yohann.de-castro (at) inria.fr


Welcome !

My main interest lies in Statistics and Machine Learning: Statistical Learning; High-Dimensional Statistics; Random Graphs; Signal Processing; Convex Optimization; Sum-Of-Squares Hierarchies; Post-Selection Inference; Gaussian Random Geometry; Kac-Rice formula;

A resume (in french) can be found here: Download PDF resume.


Research interests

Machine Learning; Statistical Learning; Compressed Sensing;
Networks; Random Graphs; Latent Space Models; Hidden Markov Models;
Convex Optimization; Sum-of-Squares Hierarchies; Sparse Deconvolution;
Post-selection Inference; Gaussian Random Geometry; Kac-Rice Formula;

High-Dimensional Statistics
  1. Restricted Isometry Constants for Gaussian and Rademacher matrices (with S. Dallaporta), submitted, 2017.
  2. Power of the Spacing test for Least-Angle Regression (with J.-M. Azaïs & S. Mourareau), Bernoulli, Volume 24 n°1 (2018), Pages 465-492.
  3. A Rice method proof of the Null-Space Property over the Grassmannian (with J.-M. Azaïs & S. Mourareau), Annales de l’Institut Henri Poincaré, Probabilités et Statistiques, 2018 (to appear).
  4. Randomized pick-freeze for sparse Sobol indices estimation in high dimension (with A. Janon), ESAIM P&S, Volume 19, 2015, Pages 725-745.
  5. Optimal designs for Lasso and Dantzig selector using Expander Codes, IEEE Transactions on Information Theory, Volume 60, Issue 11, Nov. 2014, Pages 7293-7299.
  6. A remark on the lasso and the Dantzig selector, Statistics & Probability Letters, Volume 83, Issue 1, January 2013, Pages 304-314.
Statistical Learning
  1. On Representer Theorems and Convex Regularization (with C. Boyer, A. Chambolle, V. Duval, F. de Gournay & P. Weiss), arXiv 1806.09810.
  2. Minimax Adaptive Estimation of Nonparametric Geometric Graphs (with C. Lacour and T. M. Pham Ngoc), submitted, 2017.
  3. Nonnegative Matrix Factorization with Side Information for Time Series Recovery and Prediction (with J-M Azaïs & Y. Goude & G. Hébrail & J. Mei), IEEE Transactions on Knowledge and Data Engineering, 2018 (to appear).
  4. Nonnegative Matrix Factorization for Time Series Recovery From a Few Temporal Aggregates (with J-M Azaïs & Y. Goude & G. Hébrail & J. Mei), Proceedings of the 34th International Conference on Machine Learning, PMLR 70:2382-2390, 2017.
  5. Reconstructing undirected graphs from eigenspaces (with T. Espinasse & P. Rochet), Journal of Machine Learning Research, Volume 18, Issue 51, Pages 1−24, 2017.
  6. Consistent estimation of the filtering and marginal smoothing distributions in nonparametric hidden Markov models (with É. Gassiat & S. Le Corff), IEEE Transactions on Information Theory, Volume 63, Issue 8, Aug. 2017, Pages 4758-4777.
  7. Minimax adaptive estimation of non-parametric Hidden Markov Models (with É. Gassiat & C. Lacour), Journal of Machine Learning Research, Volume 17, Issue 111, Pages 1-43, 2016. Mathematica note book and Matlab code.
  8. Estimating the transition matrix of a Markov chain observed at random times (with F. Barsotti, T. Espinasse & P. Rochet), Statistics & Probability Letters, Volume 94, November 2014, Pages 98-105.
Sparse Deconvolution, SoS Hierarchies and Gaussian Random Geometry
  1. Testing Gaussian Process with Applications to Super-Resolution (with J.-M. Azaïs & S. Mourareau), arXiv 1706.00679. A python code and Jupyter notebook are available on Github.
  2. Approximate Optimal Designs for Multivariate Polynomial Regression (with F. Gamboa & D. Henrion & R. Hess & J.-B. Lasserre), Annals of Statistics, 2018 (to appear) and its Matlab code.
    Extended version of D-Optimal Design for Multivariate Polynomial Regression via the Christoffel function and Semidefinite Relaxations (with F. Gamboa & D. Henrion & R. Hess & J.-B. Lasserre), arXiv 1703.01777.
  3. Adapting to Unknown Noise Level in Sparse Deconvolution (with C. Boyer & J. Salmon), Information & Inference: A Journal of the IMA, Volume 6, Issue 3, 1 September 2017, Pages 310–348.
  4. Exact solutions to Super Resolution on semi-algebraic domains in higher dimensions (with F. Gamboa & D. Henrion & J.-B. Lasserre), IEEE Transactions on Information Theory, Volume 63, Issue 1, Jan. 2017, Pages 621-630. Its Matlab code.
  5. Non-uniform spline recovery from small degree polynomial approximation (with G. Mijoule), Journal of Mathematical Analysis and Applications, Volume 430, Issue 2, 15 October 2015, Pages 971–992.
  6. Spike Detection from Inaccurate Samplings (with J.-M. Azaïs & F. Gamboa), Applied and Computational Harmonic Analysis, Volume 38, Issue 2, March 2015, Pages 177–195.
  7. Exact Reconstruction using Beurling Minimal Extrapolation (with F. Gamboa), Journal of Mathematical Analysis and Applications, Volume 395, Issue 1, Nov. 2012, Pages 336-354.
  8. Quantitative Isoperimetric Inequalities on the Real Line, Annales Mathématiques Blaise Pascal, Volume 18 n°2 (2011), Pages 311-331.

The complete list of my preprints can be found on HAL.

Talks and Lectures


  • Séminaire d'Informatique de l'École Normale Supérieure, Lyon.
  • Séminaire de Probabilités de l'École Normale Supérieure, Lyon.
  • Groupe de Travail "Gaussian Process" Université Jean Monnet, St-Étienne.
  • Séminaire de Probabilités de Lille.
  • Séminaire de Probabilités et Statistique de Liège.
  • Séminaire de Probabilités et Statistique de Versailles, LMV.
  • Séminaire de Statistique de Toulouse, IMT.
  • Groupe de Travail "Sequential Structured Statistical Learning", IHES.
  • Cambridge Statistics Seminar, Cambridge, UK.
  • Séminaire de Probabilités de l'École Nationale Supérieure de Techniques Avancées, Palaiseau.
  • Séminaire de Probabilités et Statistique de Nanterre, Modal'X.
  • Séminaire de Probabilités de Rennes, IRMAR.
  • 2015 Saint Flour summer school.
  • Rencontres Statistique Lyonnaises.
  • Séminaire du CREST, ENSAE.


Année universitaire 2017/2018

  • Cours et TDs de "Mise à Niveau en Mathématiques" pour les M1 BIBS, les notes de cours, les TPs et les TDs sont sur Dokeos.


  • Probabilités (MAP-PRB1) en deuxième année de l'ENSTA,
  • Statistique en M1 Mathématiques Fondamentales et Appliquées,
  • Statistique en M1 Bioinformatique et Biostatistiques,
  • Statistique en M2 Pro Ingénierie,
  • Probabilités en L2 et L3 Mathématiques,
  • Probabilités et Statistique en M1 "Biologie et Informatique",
  • Probabilités et Statistique en L2 Biologie,
  • Quatre ans colleur en classes préparatoires aux grandes écoles (environ 180h).

Habilitation à Diriger des Recherches

Le manuscrit est disponible ICI.

Ph.D. students

Jiali MEI (2014-2017 with Jean-Marc Azaïs, Yannig Goude and Georges Hébrail); "Time series recovery and prediction with regression-enhanced nonnegative matrix factorization applied to electricity consumption" defended in December 2017.

Ernesto ARAYA (Started in 2017); Random Graphs.

Jérôme Alexis CHEVALIER (Started in 2017 with Joseph Salmon and Bertrand Thirion); Statistical Learning.