Ensuring Rapid Mixing and Low Bias for Asynchronous Gibbs Sampling
This work addresses the challenge of parallelizing Gibbs sampling for faster computation, but it is incremental as it builds on existing methods to analyze asynchronous variants.
The paper tackled the problem of understanding bias and mixing time in asynchronous Gibbs sampling, showing that their theoretical results align with experimental outcomes.
Gibbs sampling is a Markov chain Monte Carlo technique commonly used for estimating marginal distributions. To speed up Gibbs sampling, there has recently been interest in parallelizing it by executing asynchronously. While empirical results suggest that many models can be efficiently sampled asynchronously, traditional Markov chain analysis does not apply to the asynchronous case, and thus asynchronous Gibbs sampling is poorly understood. In this paper, we derive a better understanding of the two main challenges of asynchronous Gibbs: bias and mixing time. We show experimentally that our theoretical results match practical outcomes.