Sample Adaptive MCMC

                     Michael H. Zhu
                 Department of Computer Science
                    Stanford University
                    Stanford, CA 94305
                  mhzhu@cs.stanford.edu

                       Abstract
     For MCMC methods like Metropolis-Hastings, tuning the proposal distribution is
     important in practice for effective sampling from the target distribution π. In this
     paper, we present Sample Adaptive MCMC (SA-MCMC), a MCMC method ...