BayesFlow Can Reliably Detect Model Misspecification and Posterior Errors in Amortized Bayesian Inference. (arXiv:2112.08866v2 [stat.ME] UPDATED)

Neural density estimators have proven remarkably powerful in performing
efficient simulation-based Bayesian inference in various research domains. In
particular, the BayesFlow framework uses a two-step approach to enable
amortized parameter estimation in settings where the likelihood function is
implicitly defined by a simulation program. But how faithful is such inference
when simulations are poor representations of reality? In this paper, we
conceptualize the types of model misspecification arising in simulation-based
inference and systematically investigate the performance of the BayesFlow
framework …

arxiv bayesian bayesian inference errors posterior