Shortened Bridge Sampler: Using deterministic approximations to accelerate SMC for posterior sampling

Sophie Donnet (1) , Stéphane Robin (1)

1 MIA-Paris - Mathématiques et Informatique Appliquées

Sequential Monte Carlo has become a standard tool for Bayesian Inference of complex models. This approach can be computationally demanding, especially when initialized from the prior distribution. On the other hand, deter-ministic approximations of the posterior distribution are often available with no theoretical guaranties. We propose a bridge sampling scheme starting from such a deterministic approximation of the posterior distribution and targeting the true one. The resulting Shortened Bridge Sampler (SBS) relies on a sequence of distributions that is determined in an adaptive way. We illustrate the robustness and the efficiency of the methodology on a large simulation study. When applied to network datasets, SBS inference leads to different statistical conclusions from the one supplied by the standard variational Bayes approximation.

Mots clés: Approximate posterior distribution; Bayesian statistics; Sequential Monte Carlo; Bridge sampling

Domaines: Applications [stat.AP]; Méthodologie [stat.ME]

Sophie Donnet, Stéphane Robin. Shortened Bridge Sampler: Using deterministic approximations to accelerate SMC for posterior sampling. 2017. ⟨hal-01566898⟩