A geometric ensemble method for Bayesian inference

Andrey A Popov

Conventional approximations to Bayesian inference rely on either approximations by statistics such as mean and covariance or by point particles. Recent advances such as the ensembl…