Paul Egels (pegels@lpsm.paris)

I do research in mathematics applied to statistics and machine learning. Currently, I am a PhD student at LPSM supervised by Prof. Ismaël Castillo.

My work focuses on high-dimensional statistics, in particular Bayesian nonparametrics, variational inference and Bayesian deep learning. I am also interested in approximation theory and other theoretical aspects of machine learning.

Papers & preprints

  1. Heavy-tailed and Horseshoe priors for regression and sparse Besov rate, with Sergios Agapiou & Ismaël Castillo (2026+), [arXiv]. (In revision for Bernoulli)
  2. Posterior and variational inference for deep neural networks with heavy-tailed weights, with Ismaël Castillo (2025), [JMLR].

Working papers

  1. Leveraging p-exp tails for adaptation, with Sergios Agapiou & Ismaël Castillo (soon available).
  2. Contraction rates of adversarially robust "pseudo"-posteriors, with Ismaël Castillo, Riccardo Lazzarini & Botond Szabo.

Talks and posters

Teaching