Variational Bayesian Monte Carlo
              with Noisy Likelihoods

                        Luigi Acerbi
                    Department of Computer Science
                      University of Helsinki
                    luigi.acerbi@helsinki.fi
                          Abstract
       Variational Bayesian Monte Carlo (VBMC) is a recently introduced framework that
       uses Gaussian process surrogates to perform approximate Bayesian inference in
       models with black-box, non-cheap lik ...