7 papers
Mixing times of data-augmentation Gibbs samplers for high-dimensional probit regression
Filippo Ascolani, Giacomo Zanella
We investigate the convergence properties of popular data-augmentation samplers for Baye\-sian probit regression. Leveraging recent results on Gibbs samplers for log-concave target…
Entropy contraction of the Gibbs sampler under log-concavity
Filippo Ascolani, Hugo Lavenant, Giacomo Zanella
The Gibbs sampler (a.k.a. Glauber dynamics and heat-bath algorithm) is a popular Markov Chain Monte Carlo algorithm which iteratively samples from the conditional distributions of…
Scalability of Metropolis-within-Gibbs schemes for high-dimensional Bayesian models
Filippo Ascolani, Gareth O. Roberts, Giacomo Zanella
We study general coordinate-wise MCMC schemes (such as Metropolis-within-Gibbs samplers), which are commonly used to fit Bayesian non-conjugate hierarchical models. We relate their…
Error Bounds and Optimal Schedules for Masked Diffusions with Factorized Approximations
Hugo Lavenant, Giacomo Zanella
Recently proposed generative models for discrete data, such as Masked Diffusion Models (MDMs), exploit conditional independence approximations to reduce the computational cost of p…
A fast non-reversible sampler for Bayesian finite mixture models
Filippo Ascolani, Giacomo Zanella
Finite mixtures are a cornerstone of Bayesian modelling, and it is well-known that sampling from the resulting posterior distribution can be a hard task. In particular, popular rev…
Convergence rate of random scan Coordinate Ascent Variational Inference under log-concavity
Hugo Lavenant, Giacomo Zanella
The Coordinate Ascent Variational Inference scheme is a popular algorithm used to compute the mean-field approximation of a probability distribution of interest. We analyze its ran…