Output-Sensitive Adaptive Metropolis-Hastings for Probabilistic Programs
نویسندگان
چکیده
We introduce an adaptive output-sensitive Metropolis-Hastings algorithm for probabilistic models expressed as programs, Adaptive Lightweight Metropolis-Hastings (AdLMH). The algorithm extends Lightweight Metropolis-Hastings (LMH) by adjusting the probabilities of proposing random variables for modification to improve convergence of the program output. We show that AdLMH converges to the correct equilibrium distribution and compare convergence of AdLMH to that of LMH on several test problems to highlight different aspects of the adaptation scheme. We observe consistent improvement in convergence on the test problems.
منابع مشابه
Learning Stochastic Inverses for Adaptive Inference in Probabilistic Programs
We describe an algorithm for adaptive inference in probabilistic programs. During sampling, the algorithm accumulates information about the local probability distributions that compose the program’s overall distribution. We use this information to construct targeted samples: given a value for an intermediate expression, we stochastically invert each of the steps giving rise to this value, sampl...
متن کاملSemantics Sensitive Sampling for Probabilistic Programs
We present a new semantics sensitive sampling algorithm for probabilistic programs, which are “usual” programs endowed with statements to sample from distributions, and condition executions based on observations. Since probabilistic programs are executable, sampling can be performed by repeatedly executing them. However, in the case of programs with a large number of random variables and observ...
متن کاملAdaptive Scheduling in MCMC and Probabilistic Programming
We introduce an adaptive output-sensitive inference algorithm for MCMC and probabilistic programming, Adaptive Random Database. The algorithm is based on a single-site updating Metropolis-Hasting sampler, the Random Database (RDB) algorithm. Adaptive RDB (ARDB) differs from the original RDB in that the schedule of selecting variables proposed for modification is adapted based on the output of o...
متن کاملOn the ergodicity properties of someadaptive MCMC algorithms
In this paper we study the ergodicity properties of some adaptive Markov chain Monte Carlo algorithms (MCMC) that have been recently proposed in the literature. We prove that under a set of verifiable conditions, ergodic averages calculated from the output of a so-called adaptive MCMC sampler converge to the required value and can even, under more stringent assumptions, satisfy a central limit ...
متن کاملGenerating Efficient MCMC Kernels from Probabilistic Programs
Universal probabilistic programming languages (such as Church [6]) trade performance for abstraction: any model can be represented compactly as an arbitrary stochastic computation, but costly online analyses are required for inference. We present a technique that recovers hand-coded levels of performance from a universal probabilistic language, for the Metropolis-Hastings (MH) MCMC inference al...
متن کامل