Presentation

I am an assistant professor in statistics at ENSAI (Rennes) working on topics related to machine learning and high-dimensional statistics. My PhD centered on two areas in modern statistics: change-point detection and ranking problems.

I am interested in GPU parallel computing. I’m currently developing a Julia library focused on optimizing fundamental functions like mapreduce (for operations such as sum) and accumulate (for operations like prefix sum). You can explore my experimental work at my GitHub repository Luma for more technical details.

contact: firstname.lastname@ensai.fr

Publications

  • E. Pilliat, A. Carpentier, N. Verzelen, Optimal rates for ranking a permuted isotonic matrix in polynomial time (2024) SODA
  • E. Pilliat, A. Carpentier, N. Verzelen, Optimal permutation estimation in crowd-sourcing problems (2023) Annals of Statistics. [arxiv], [presentation],[poster]
  • E. Pilliat, A. Carpentier, N. Verzelen, Optimal multiple change-point detection for high-dimensional data (2023) EJS [arxiv]

Vitae

  • Research

    • Sep 2024 - Present: assistant professor in statistics at ENSAI
    • 2024: Postdoctoral researcher at ENS Lyon
    • Feb 2020 - Apr 2022 and Oct 2022 - Dec 2023: PhD Student at Université de Montpellier and INRAE
    • 2019: Research Project at Ecole Normale Supérieure Paris Saclay with Vianney Perchet on Optimal Order Selection for an Online Reward Maximization problem
    • Apr 2019: Research Project at OVGU Magdeburg with Alexandra Carpentier on Signal Detection and Change-Point Detection
    • 2018: Research Internship at the University of Cambridge on Gaussian Free Field

  • Experience

    • Apr 2022 - Oct 2022: Quantitative Intern at QRT

  • Education

    • 2016-2020: Student at Ecole Normale Supérieure de Lyon
    • 2013-2016: Preparatory School of Mathematics and Physics (CPGE MPSI/MP*) in Strasbourg

  • Teaching

    • 2020-2022: Teaching assistant at Université de Montpellier.