Thibault Randrianarisoa

Thibault Randrianarisoa

Postdoctoral researcher

Bocconi University

About me

I am a Postdoctoral research at BIDSA at Bocconi University (Milan, Italy) working with Botond Szabo. I hold a PhD in Statistics from Sorbonne Université (Paris, France), where I worked in the LPSM lab under the supervision of Ismaël Castillo. I am mainly interested in high-dimensional and nonparametric statistics. My PhD manuscript can be found here.

I am/have been a member of the following projects:

Interests

  • Bayesian nonparametrics
  • (Deep) Gaussian processes
  • Tree-based methods (Random Forests, CART, Pólya Trees, BART,…)
  • Uncertainty quantification
  • Variational Bayes
  • Inverse problems
  • Differential privacy

Education

  • PhD in Statistics, 2019-2022

    Sorbonne Université

  • StatML Master Degree, 2018-2019

    Université Paris-Saclay

  • MSc in Statistics and Data Science, 2015-2019

    ENSAE- Institut Polytechnique de Paris

Publications

Quickly discover relevant content by filtering publications.

Deep Horseshoe Gaussian Processes

Submitted
PDF
Deep Horseshoe Gaussian Processes

Variational Gaussian processes for linear inverse problems

Advances in Neural Information Processing Systems 36 (NeurIPS 2023)
PDF
Variational Gaussian processes for linear inverse problems

On Adaptive Confidence Sets for the Wasserstein Distances

Bernoulli, 2023
PDF DOI

Optional Pólya trees: posterior rates and uncertainty quantification

Electronic Journal of Statistics, 2022
PDF DOI
Optional Pólya trees: posterior rates and uncertainty quantification

Smoothing and adaptation of shifted Pólya Tree ensembles

Bernoulli, 2020
PDF DOI

Recent & Upcoming Talks

Contributions to the theoretical analysis of statistical learning and uncertainty quantification methods

My PhD defence will be held on the 28th of September, 2pm, in room 16-26-209 (Paul Lévy) of Campus Pierre et Marie Curie (Sorbonne Université).
Sep 28, 2022 14:00 — 18:00 Sorbonne Université (Campus Pierre et Marie Curie), room 16-26-209 (Paul Lévy)

Optional Pólya trees: vitesses de contraction de la loi a posteriori et quantification de l’erreur

Aug 26, 2021 14:30 — 15:00

Smoothing and adaptation of shifted Pólya Tree ensembles

Jun 30, 2021 14:15
Slides

A toy model of Polya tree ensemble: smoothing and adaptation

Jul 15, 2020 11:00 — 11:40
Slides Video

Smoothing of Bayesian forest estimators in density estimation

With CART regression trees, the values of the estimator in the leaf cells are equal to the mean of the observed data in each cell. …
Jun 8, 2020 17:00 — 18:00

Communication

For the 2020-2021 academic year, I co-organized the LPSM weekly PhD student seminar (GTT)

Teaching

I was a co-instructor of the following course:

2023-2024

  • Mathematical Statistics (BSc, Bocconi University) with Botond Szabó.

I have been a teaching assistant for the following courses:

2020-2021

2019-2020

Experience

 
 
 
 
 

Machine Learning Research Intern

Walnut Algorithms

Mar 2018 – Sep 2018 Paris
Design of a trading algorithm.
 
 
 
 
 

Software Engineering Intern

Spinergie

Sep 2017 – Feb 2018 Paris
Built a large-scale crawler/scraper in Python.
 
 
 
 
 

Biostatistician Intern

Institut de Recherches Internationales SERVIER (I.R.I.S.)

Jun 2016 – Sep 2016 Paris
Review of statistical methods associated with interim analyses of clinical trials.

Contact

  • via Roentgen 1
    20136 Milano
  • My office is located on the 3rd Floor of Tower 15/16 and is at number 3-C3-03.
  • Book an appointment