E
Eric Moulines
Researcher at Institut Mines-Télécom
Publications - 16
Citations - 408
Eric Moulines is an academic researcher from Institut Mines-Télécom. The author has contributed to research in topics: Matrix completion & Particle filter. The author has an hindex of 8, co-authored 16 publications receiving 309 citations. Previous affiliations of Eric Moulines include École Polytechnique.
Papers
More filters
Journal ArticleDOI
An Adaptive Parallel Tempering Algorithm
TL;DR: In this paper, an adaptive algorithm with fixed number of temperatures is proposed, which tunes both the temperature schedule and the parameters of the random-walk Metropolis kernel automatically, and proves the convergence of the adaptation and a strong law of large numbers for the algorithm under general conditions.
Posted Content
Non-asymptotic Analysis of Biased Stochastic Approximation Scheme
TL;DR: This work analyzes a general SA scheme to minimize a non-convex, smooth objective function, and illustrates these settings with the online EM algorithm and the policy-gradient method for average reward maximization in reinforcement learning.
Journal ArticleDOI
Quantitative bounds of convergence for geometrically ergodic Markov chain in the Wasserstein distance with application to the Metropolis Adjusted Langevin Algorithm
Alain Durmus,Eric Moulines +1 more
TL;DR: The proposed rate of convergence leads to useful insights for the analysis of MCMC algorithms, and suggests ways to construct sampler with good mixing rate even if the dimension of the underlying sampling space is large.
Journal ArticleDOI
Adaptive Multinomial Matrix Completion
TL;DR: In this article, the authors investigated the case of highly quantized observations when the measurements can take only a small number of values and proposed a constrained, nuclear norm penalized maximum likelihood estimator.
Journal ArticleDOI
Uniform Ergodicity of the Particle Gibbs Sampler
TL;DR: The particle Gibbs sampler is a systematic way of using a particle filter within Markov chain Monte Carlo as discussed by the authors, which results in an off-the-shelf Markov kernel on the space of state trajectories, which can be used to simulate from the full joint smoothing distribution for a state space model.