Accelerating Metropolis-Hastings with Lightweight Inference Compilation
arXiv:2010.12128
Abstract
In order to construct accurate proposers for Metropolis-Hastings Markov Chain Monte Carlo, we integrate ideas from probabilistic graphical models and neural networks in an open-source framework we call Lightweight Inference Compilation (LIC). LIC implements amortized inference within an open-universe declarative probabilistic programming language (PPL). Graph neural networks are used to parameterize proposal distributions as functions of Markov blankets, which during "compilation" are optimized to approximate single-site Gibbs sampling distributions. Unlike prior work in inference compilation (IC), LIC forgoes importance sampling of linear execution traces in favor of operating directly on Bayesian networks. Through using a declarative PPL, the Markov blankets of nodes (which may be non-static) are queried at inference-time to produce proposers Experimental results show LIC can produce proposers which have less parameters, greater robustness to nuisance random variables, and improved posterior sampling in a Bayesian logistic regression and -schools inference application.
References in corpus (9)
- Semi-Supervised Classification with Graph Convolutional Networks
- Spectral Networks and Locally Connected Networks on Graphs
- Discriminative Embeddings of Latent Variable Models for Structured Data
- Elliptical slice sampling
- Deep Amortized Inference for Probabilistic Programs
- Iterative Amortized Inference
- Inference Compilation and Universal Probabilistic Programming
- C3: Lightweight Incrementalized MCMC for Probabilistic Programs using Continuations and Callsite Caching
- Newtonian Monte Carlo: single-site MCMC meets second-order gradient methods