Metropolis–Hastings via Classification
From MaRDI portal
Abstract: This paper develops a Bayesian computational platform at the interface between posterior sampling and optimization in models whose marginal likelihoods are difficult to evaluate. Inspired by adversarial optimization, namely Generative Adversarial Networks (GAN), we reframe the likelihood function estimation problem as a classification problem. Pitting a Generator, who simulates fake data, against a Classifier, who tries to distinguish them from the real data, one obtains likelihood (ratio) estimators which can be plugged into the Metropolis-Hastings algorithm. The resulting Markov chains generate, at a steady state, samples from an approximate posterior whose asymptotic properties we characterize. Drawing upon connections with empirical Bayes and Bayesian mis-specification, we quantify the convergence rate in terms of the contraction speed of the actual posterior and the convergence rate of the Classifier. Asymptotic normality results are also provided which justify inferential potential of our approach. We illustrate the usefulness of our approach on examples which have posed a challenge for existing Bayesian likelihood-free approaches.
Cites work
- A closed-form solution for options with stochastic volatility with applications to bond and currency options
- A comparative review of dimension reduction methods in approximate Bayesian computation
- A theory of the term structure of interest rates
- Adapting the ABC distance function
- Analyses of Infectious Disease Data from Household Outbreaks by Markov Chain Monte Carlo Methods
- Approximate Bayesian Computation with the Wasserstein Distance
- Approximate Bayesian computation: a nonparametric perspective
- Asymptotic properties of approximate Bayesian computation
- Asymptotic Statistics
- Bayesian estimation of the discrepancy with misspecified parametric models
- Bayesian inference for irreducible diffusion processes using the pseudo-marginal approach
- Bayesian optimization for likelihood-free inference of simulator-based statistical models
- Bayesian Synthetic Likelihood
- Componentwise approximate Bayesian computation via Gibbs-like steps
- Convergence properties of pseudo-marginal Markov chain Monte Carlo algorithms
- Convergence rates of posterior distributions for non iid observations
- Convergence rates of posterior distributions.
- Convex models, MLE and misspecification
- Density ratio estimation in machine learning. Foreword by Thomas G. Dietterich
- Efficient implementation of Markov chain Monte Carlo when using an unbiased likelihood estimator
- scientific article; zbMATH DE number 2117879 (Why is no real title available?)
- scientific article; zbMATH DE number 3894273 (Why is no real title available?)
- scientific article; zbMATH DE number 6276120 (Why is no real title available?)
- Learning Summary Statistic for Approximate Bayesian Computation via Deep Neural Network
- Likelihood-free inference via classification
- Maximum Likelihood Estimation of Discretely Sampled Diffusions: A Closed-form Approximation Approach
- Misspecification in infinite-dimensional Bayesian statistics
- Model Misspecification in Approximate Bayesian Computation: Consequences and Diagnostics
- On inference for partially observed nonlinear diffusion models using the Metropolis-Hastings algorithm
- On the convergence of the Metropolis-Hastings Markov chains
- Particle Markov chain Monte Carlo for efficient numerical simulation
- Relevant statistics for Bayesian model choice
- Robust Approximate Bayesian Inference With Synthetic Likelihood
- Stochastic modelling for systems biology.
- The correlated pseudomarginal method
- The pseudo-marginal approach for efficient Monte Carlo computations
- Weak convergence and empirical processes. With applications to statistics
This page was built for publication: Metropolis–Hastings via Classification
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6144768)
