Asynchronous Gibbs Sampling
arXiv:1509.08999
Abstract
Gibbs sampling is a Markov Chain Monte Carlo (MCMC) method often used in Bayesian learning. MCMC methods can be difficult to deploy on parallel and distributed systems due to their inherently sequential nature. We study asynchronous Gibbs sampling, which achieves parallelism by simply ignoring sequential requirements. This method has been shown to produce good empirical results for some hierarchical models, and is popular in the topic modeling community, but was also shown to diverge for other targets. We introduce a theoretical framework for analyzing asynchronous Gibbs sampling and other extensions of MCMC that do not possess the Markov property. We prove that asynchronous Gibbs can be modified so that it converges under appropriate regularity conditions -- we call this the exact asynchronous Gibbs algorithm. We study asynchronous Gibbs on a set of examples by comparing the exact and approximate algorithms, including two where it works well, and one where it fails dramatically. We conclude with a set of heuristics to describe settings where the algorithm can be effectively used.
References in corpus (3)
Cited by in corpus (11)
- Streaming Graph Challenge: Stochastic Block Partition
- Decentralized Gaussian Filters for Cooperative Self-localization and Multi-target Tracking
- GPU-accelerated Gibbs sampling: a case study of the Horseshoe Probit model
- Communication-Efficient Distributed Statistical Inference
- Ensuring Rapid Mixing and Low Bias for Asynchronous Gibbs Sampling
- Pólya Urn Latent Dirichlet Allocation: a doubly sparse massively parallel sampler
- Federated Stochastic Gradient Langevin Dynamics
- Anytime Monte Carlo
- Privacy-Preserving Multiple Tensor Factorization for Synthesizing Large-Scale Location Traces with Cluster-Specific Features
- Techniques for proving Asynchronous Convergence results for Markov Chain Monte Carlo methods
- Variational Gibbs Inference for Statistical Model Estimation from Incomplete Data