Highly Influential

@article{Seita2015FastPS, title={Fast Parallel SAME Gibbs Sampling on General Discrete Bayesian Networks}, author={Daniel Seita and Haoyu Chen and John F. Canny}, journal={CoRR}, year={2015}, volume={abs/1511.06416} }

- Published 2015 in ArXiv

A fundamental task in machine learning and related fields is to perform inference on Bayesian networks. Since exact inference takes exponential time in general, a variety of approximate methods are used. Gibbs sampling is one of the most accurate approaches and provides unbiased samples from the posterior but it has historically been too expensive for large models. In this paper, we present an optimized, parallel Gibbs sampler augmented with state replication (SAME or State Augmented Marginal… CONTINUE READING