Recent publications

More Publications

(2022). A pseudo-marginal sequential Monte Carlo online smoothing algorithm. Bernoulli, volume 28, number 4, p. 2606-2633.

Preprint Source Document

(2022). Diffusion bridges vector quantized variational autoencoders. Proceedings of the 39th International Conference on Machine Learning (ICML), PMLR 162:4141-4156.

Preprint Source Document

(2022). Deconvolution with unknown noise distribution is possible for multivariate signals. The Annals of Statistics, volume 50, number 1, p. 303-323.

Preprint PDF Source Document

(2022). On last layer state space models. Submitted.

(2021). Disentangling Identifiable Features from Noisy Data with Structured Nonlinear ICA. Advances in Neural Information Processing Systems 34 (NeurIPS 2021), volume 34, p. 1624-1633.

Preprint Source Document

(2021). NEO: Non Equilibrium Sampling on the Orbit of a Deterministic Transform. Advances in Neural Information Processing Systems 34 (NeurIPS 2021), volume 34, p. 17060-17071.

Preprint Source Document

(2021). Learning Natural Language Generation from Scratch. To appear in the Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL 2022).

Preprint

(2021). End-to-end deep meta modelling to calibrate and optimize energy consumption and comfort. Energy and Buildings, 250:1.

Preprint Source Document

(2021). Joint self-supervised blind denoising and noise estimation. Submitted.

Preprint

(2020). Learning the distribution of latent variables in paired comparison models with Round-Robin scheduling. Bernoulli, volume 26, number 4, p. 2670-2698.

Preprint Source Document

(2020). The Monte Carlo Transformer: a stochastic self-attention model for sequence prediction. Submitted.

Preprint

(2020). Backward importance sampling for online estimation of state space models. Submitted.

Preprint

(2020). Identifiability and consistent estimation of nonparametric translation hidden Markov models with general state space. The Journal of Machine Learning Research (JMLR), (115):1-40, 2020.

Preprint PDF Source Document

Research activities

Current Ph.D. students

  • Alice Martin (Telecom SudParis & Ecole Polytechnique, Apr. 2019 - June 2022)
    Co-supervised with Olivier Pietquin (Google Brain)
    Deep reinforcement learning for natural language processing. Application to goal-oriented visual question generation
  • Max Cohen (Telecom SudParis & Oze Energies, Feb. 2020 - …)
    Co-supervised with Marius Preda (Telecom SudParis)
    Metamodels and Bayesian deep learning for partially observed dynamical systems
  • Yazid Janati (Telecom SudParis, Oct. 2020 - …)
    Co-supervised with Yohan Petetin (Telecom SudParis)
    On the combination of deep learning models with Monte Carlo methods
  • Mathis Chagneux (Telecom Paris, Nov. 2020 - …)
    Co-supervised with Pierre gloaguen (Agro ParisTech) and Charles Ollion (Ecole Polytechnique)
    Computer vision and statistical models for the detection and tracking of plastic wastes
  • Etienne David (Telecom SudParis & Heuritech, Feb. 2021 - …)
    Co-supervised with Jean Bellot (Heuritech)
    Hybrid models and weak signals to forecast fashion trends


#4 Hi! PARIS workshop “AI for the energy transition” (January 13, 2022)
Co-organized with Anne-Laure Sellier for the Hi! PARIS center

#2 Data sharing winter school “Next Mobility – Smart Data Sharing to Move Goods and People” (December 7th - 9th, 2021)
Co-organized with Technische Universitat Dortmund, International Data Spaces Association, the German-French Academy for the Industry of the Future and the Universite franco-allemande.
Program and Additional information

#3 Hi! PARIS workshop “AI for sustainability” (October 1, 2021)
Co-organized with Anne-Laure Sellier for the Hi! PARIS center

CIRM Winter school “End-to-end Bayesian learning” (25-29th of October, 2021)
Co-chair with Pierre Gloaguen and Julien Stoehr.
Website

GdR ISIS “Apprentissage profond et modeles generatifs pour modeliser l’incertitude des donnees (Session 2)” (17 May 2021)
Co-organized with Francois Septier. Program

#2 Hi! PARIS workshop “AI for healthcare” (April 9, 2021)
Co-organized with Anne-Laure Sellier for the Hi! PARIS center

#1 workshop of the chair Interpretable AI for mission-critical applications with Devoteam
Roles et impacts de l’IA dans les transformations de l’industrie du retail (April 21st, January )
Co-organized with Cecile Romain

#1 Hi! PARIS workshop “AI bias and Data Privacy” (January 15, 2021)
Co-organized with Anne-Laure Sellier for the Hi! PARIS center

#1 Data sharing winter school “Data sharing in the manufacturing industry - deepening knowledge to build data spaces” (December 2nd - 4th, 2020)
Co-organized with Technische Universitat Dortmund, International Data Spaces Association, the German-French Academy for the Industry of the Future and the Universite franco-allemande.
Program and Additional information

GdR ISIS “Apprentissage profond et modeles generatifs pour modeliser l’incertitude des donnees (Session 1)” (11 May 2020)
Co-organized with Francois Septier. Program and Slides

Monthly seminar “All about that Bayes” Seminar website
Co-organized with Pierre Gloaguen, Julien Stoehr and Alain Durmus

“2019 Workshop on machine learning at Telecom SudParis” (15 November 2019)
Co-organized with Randal Douc. Registration (free registration required) and Program

Co-chair of the Junior Conference on Data Science and Engineering JDSE20 (12 & 13 September 2019)
Chair with Gianluca Quercini and Isabelle Huteau

Co-chairholder Data science pour le e-commerce with Telecom ParisTech and VeePee
Official opening (17 April 2019)

Ateliers de statistiques de la SFDS 2019 (09-11 September 2019)
Des MCMC aux nouveaux algorithmes d’inference bayesienne : HMC, HMC/SMC, Variational Bayes, ABC…

Algorithms & Simulations

Notebooks, Matlab codes, R Markdown associated with scientific publications

Data sciences material

Publications about working group sessions at TSP

Students

Supervision activities (internships, Ms.Sc., Ph.D., fellowships)

Teaching

M2 Machine Learning
Ms. Sc. Data Sciences (2nd year / M2) @ISUP

Lecture 1: introduction SlidesExercisesAdditional notes
Lecture 2: Discriminant Analysis SlidesExercises
Lecture 3: Logistic regression SlidesExercises



M1 Statistical learning
Ms. Sc. Data Sciences (1st year / M1) @ISUP

Lecture notes (as of 18/09/2022) Notes
Lecture 1: Introduction (Chapter 1 - Sections 1.1 & 1.2 and Chapter 2 - Section 2.1) Exercises
Lecture 2: Full rank linear regression (Chapter 2 - Sections 2.2 to 2.4) ExercisesA few simulations [Brazilian inflation]
Lecture 3: Ridge regression (Chapter 3 - Section 3.1) A few simulations [Follow-up Brazilian inflation]







Lecture notes

Introduction to Markov chains and Markov chain Monte Carlo methods.
Ms. Sc. Data Sciences (2nd year / M2) @Institut Polytechnique de Paris

Introduction to machine learning.
Ms. Eng. (2nd year) @Institut Polytechnique de Paris
Dimension reduction, supervised classification, multilinear regression, gradient based optimization…

Bayesian inference for partially observed dynamical systems
Ms. Sc. Data Sciences (2nd year / M2) @Institut Polytechnique de Paris

Markov chains Exercises

Maximum likelihood estimation for Markov chains Exercises

Markov Chain Monte Carlo methods Exercises

Asymptotic properties of Markov chains Exercises

Metropolis-Hastings, MALA, Hamiltonian Monte Carlo Practical session (Python ipynb)

Introduction to machine learning Ms. Eng. (2nd year) @Institut Polytechnique de Paris

Dimension reduction: Principal component analysis, independent component analysis
Slides and Practical session (Python ipynb)

Supervised classification (I): Bayes classifier, discriminant analysis, Support vector machines
Slides and Practical session (Python ipynb)

Supervised classification (II): Logistic regression, Feed forward neural networks
Slides and Practical session (Python ipynb)

Gradient descent based optimization algorithms
Slides and Practical session (Python ipynb)

Kernel based regression and random forests
Practical session (Python ipynb)

Introduction to computational statistics, machine learning and deep learning

Ateliers SFDS 2019: MCMC and introduction to Hamiltonian Monte Carlo HMC, ABC and variational (Python ipynb)

Dimension reduction (SVD, PCA, use before machine learning algorithms) PCA (Python ipynb).

Introduction to machine learning (random forest, gradient descent, feed forward neural networks) Intro to machine learning (Python ipynb).

Machine learning algorithms for the Ms. Sc. Advanced Machine learning (3rd year) @Ecole Polytechnique

AdaBoost and Random forests pdf.

Kernel Principal Component Analysis pdf.

Kmeans and Expectation Maximization algorithms pdf.

Machine learning algorithms for the Ms. Sc. Machine learning (2nd year) @TSP

Singular value decomposition for image reconstruction svd (Python ipynb).

Linear and quadratic discriminant analysis lda (R Markdown).

Support Vector Machines - application to well being at work data svm (Python ipynb).

Simulation of stochatic differential equations SDE (Python ipynb).

Advanced Statistics exercises for the Ms. Sc. MA2822 (2nd year) @Centrale Supelec

Introduction to simulation and Monte Carlo methods pdf.

Kernel based density estimators pdf.

Stochastic calculus Ms. Sc. Machine learning (2nd year) @TSP

Lecture notes on stochastic differential equations (v0) pdf.