🤖 AI Summary
Analytic computation of reference priors—particularly Jeffreys priors—is intractable in small-sample Bayesian inference. Method: This paper proposes a neural variational inference framework that automatically learns parameterized approximations to reference priors. It is the first approach to integrate variational inference with deep neural networks for objective prior approximation; it enforces prior propriety via constrained optimization and enables efficient posterior recovery via MCMC even when the prior density is analytically unavailable. Contribution/Results: Experiments across diverse statistical models demonstrate successful reconstruction of theoretical reference priors, joint optimization of prior and posterior, and substantial improvements in objectivity and accuracy of Bayesian estimation under small-sample regimes.
📝 Abstract
In Bayesian statistics, the choice of the prior can have an important influence on the posterior and the parameter estimation, especially when few data samples are available. To limit the added subjectivity from a priori information, one can use the framework of reference priors. However, computing such priors is a difficult task in general. We develop in this paper a flexible algorithm based on variational inference which computes approximations of reference priors from a set of parametric distributions using neural networks. We also show that our algorithm can retrieve reference priors when constraints are specified in the optimization problem to ensure the solution is proper. We propose a simple method to recover a relevant approximation of the parametric posterior distribution using Markov Chain Monte Carlo (MCMC) methods even if the density function of the parametric prior is not known in general. Numerical experiments on several statistical models of increasing complexity are presented. We show the usefulness of this approach by recovering the target distribution. The performance of the algorithm is evaluated on the prior distributions as well as the posterior distributions, jointly using variational inference and MCMC sampling.