Zobrazeno 1 - 10
of 333
pro vyhledávání: '"Guillin, Arnaud"'
In this paper, we rigorously derive Central Limit Theorems (CLT) for Bayesian two-layerneural networks in the infinite-width limit and trained by variational inference on a regression task. The different networks are trained via different maximizatio
Externí odkaz:
http://arxiv.org/abs/2406.09048
In this paper, we prove in a very weak regularity setting existence and uniqueness of quasi-stationary distributions as well as exponential convergence towards the quasi-stationary distribution for the generalized Langevin and the Nos{\'e}-Hoover pro
Externí odkaz:
http://arxiv.org/abs/2403.17471
In this article, we focus on two toy models : the Curie-Weiss model and the system of $N$ particles in linear interactions in a double well confining potential. Both models, which have been extensively studied, describe a large system of particles wi
Externí odkaz:
http://arxiv.org/abs/2308.01624
We provide a rigorous analysis of training by variational inference (VI) of Bayesian neural networks in the two-layer and infinite-width case. We consider a regression problem with a regularized evidence lower bound (ELBO) which is decomposed into th
Externí odkaz:
http://arxiv.org/abs/2307.04779
Event-Chain Monte Carlo methods generate continuous-time and non-reversible Markov processes which often display important accelerations compared to their reversible counterparts. However their generalization to any system may appear less straightfor
Externí odkaz:
http://arxiv.org/abs/2307.02341
We consider a general system of two run-and-tumble particles interacting by hardcore jamming on the unidimensional torus. RTP are a paradigmatic active matter model, typically modeling the evolution of bacteria. By directly modeling the system at the
Externí odkaz:
http://arxiv.org/abs/2306.00831
Normalizing Flows (NF) are Generative models which are particularly robust and allow for exact sampling of the learned distribution. They however require the design of an invertible mapping, whose Jacobian determinant has to be computable. Recently i
Externí odkaz:
http://arxiv.org/abs/2302.01955
Monte Carlo simulations of systems of particles such as hard spheres or soft spheres with singular kernels can display around a phase transition prohibitively long convergence times when using traditional Hasting-Metropolis reversible schemes. Effici
Externí odkaz:
http://arxiv.org/abs/2208.11070
In this work, we consider a wide two-layer neural network and study the behavior of its empirical weights under a dynamics set by a stochastic gradient descent along the quadratic loss with mini-batches and noise. Our goal is to prove a trajectorial
Externí odkaz:
http://arxiv.org/abs/2207.12734
In this article, we prove the first quantitative uniform in time propagation of chaos for a class of systems of particles in singular repulsive interaction in dimension one that contains the Dyson Brownian motion. We start by establishing existence a
Externí odkaz:
http://arxiv.org/abs/2204.10653