Scalable Bernoulli factories for Bayesian inference with intractable likelihoods

Abstract: Bernoulli factory MCMC algorithms implement accept-reject Markov chains without explicit computation of acceptance probabilities, and are used to target posterior distributions associated with intractable likelihood models. These algorithms often mix better than alternatives based on data augmentation or acceptance probability estimation. However, we show that their computational performance typically deteriorates exponentially with data size. To address this, we propose a simple divide-and-conquer Bernoulli factory MCMC algorithm and prove that it has polynomial complexity of degree between 1 and 2, with the exact degree depending on the existence of efficient unbiased estimators of the intractable likelihood ratio. We demonstrate the effectiveness of our approach with applications to Bayesian inference in two intractable likelihood models, and observe respective polynomial cost of degree 1.2 and 1 in the data size.