Poisson-Minibatching for Gibbs Sampling with
       Convergence Rate Guarantees

          Ruqi Zhang               Christopher De Sa
         Cornell University             Cornell University
        rz297@cornell.edu             cdesa@cs.cornell.edu

                       Abstract
     Gibbs sampling is a Markov chain Monte Carlo method that is often used for learn-
     ing and inference on graphical models. Minibatching, in which a small random
     subset of the graph is used at each itera ...