Talk Title

Constraining the Geometric Rate of Convergence of Random Walk Metropolis-Hastings

Abstract

Convergence rate analyses of Metropolis-Hastings Markov chains on general state spaces have largely focused on establishing sufficient conditions for geometric ergodicity without providing explicit constraints on the rate of convergence.  Constraints are provided for random walk Metropolis-Hastings Markov chains by developing upper and lower bounds on the geometric rate of convergence.  The results are applied to a large class of exponential families and generalized linear models that address Bayesian Regression problems.

Bio

I am a Post-Doctoral Fellow at Purdue University. I have graduated from the University of Minnesota with a PhD in statistics. My research interests include Reinforcement Learning, Markov Chain Monte Carlo Methods, Stochastic Optimization and Machine Learning.