Joseph Lam-Weil

Joseph Lam-Weil

Theoretical Machine Learning, Statistical testing, Bandit, medical applications.

I am a postdoc at Lorraine University in Nancy, with Anne Gégout-Petit, Denis Villemonais and Simon Toupance since April 2021.

Publications

• A minimax near-optimal algorithm for adaptive rejection sampling, ALT 2019
  Juliette Achdou, JL-W, Alexandra Carpentier, Gilles Blanchard.
  Code

• Local minimax rates for closeness testing of discrete distributions, Bernoulli 2022
  JL-W, Alexandra Carpentier, Bharath K. Sriperumbudur.
  

• Minimax optimal goodness-of-fit testing for densities under a local differential privacy constraint, Bernoulli 2022
  JL-W, Béatrice Laurent, Jean-Michel Loubes.
  

Experience

PhD

• Magdeburg university (November 2017-March 2021): under the supervision of Alexandra Carpentier.

Visiting student

• Paul Sabatier university of Toulouse (November 2018-December 2018): worked on identity testing under local differential privacy constraint, with Jean-Michel Loubes and Beatrice Laurent-Bonneau.

Research intern

• Thales (May 2017-October 2017): worked on anomaly detection using deep learning methods, under the supervision of Marc Schoenauer.

• xbird (June 2016-September 2016): applied machine learning methods to glycemia prediction for diabetics.

Teaching Assistant

• Advanced Probability Theory: 1 semester.

• Introduction to Probability and Statistics: 2 semesters.

• Probability Theory: 1 semester.

Events

• MLSS 2019 in London.