P
Pierre Alquier
Researcher at Université Paris-Saclay
Publications - 97
Citations - 1949
Pierre Alquier is an academic researcher from Université Paris-Saclay. The author has contributed to research in topics: Estimator & Bayesian probability. The author has an hindex of 23, co-authored 97 publications receiving 1597 citations. Previous affiliations of Pierre Alquier include ENSAE ParisTech & University College Dublin.
Papers
More filters
Journal ArticleDOI
Noisy Monte Carlo: convergence of Markov chains with approximate transition kernels
TL;DR: In this article, the authors explore a variety of situations where it is possible to quantify how close the chain given by the transition kernel given by a Markov chain is to the chain generated by a transition kernel.
Journal Article
On the properties of variational approximations of Gibbs posteriors
TL;DR: In this paper, the authors consider variational approximations of the Gibbs posterior, which are fast to compute and have the same rate of convergence as the original PAC-Bayesian procedure.
Journal ArticleDOI
Model selection for weakly dependent time series forecasting
TL;DR: In this article, the authors propose a two-step procedure for predicting the next value of a stationary time series, where the first step follows machine learning theory paradigm and consists in determining a set of possible predictors as randomized estimators in (possibly numerous) different predictive models.
Journal ArticleDOI
Concentration of tempered posteriors and of their variational approximations
Pierre Alquier,James Ridgway +1 more
TL;DR: A general approach to prove the concentration of variational approximations of fractional posteriors of matrix completion and Gaussian VB is proposed.
Posted Content
Sparse single-index model
Pierre Alquier,Gérard Biau +1 more
TL;DR: This work considers the single-index model estimation problem from a sparsity perspective using a PAC-Bayesian approach and offers a sharp oracle inequality, which is more powerful than the best known oracle inequalities for other common procedures of single- index recovery.