UBC Probability Seminar: Jeffrey Rosenthal

  • Date: 02/22/2024
  • Time: 11:00
Jeffrey Rosenthal, University of Toronto

University of British Columbia


Speeding up Metropolis using Theorems


Markov chain Monte Carlo (MCMC) algorithms, such as the Metropolis algorithm, are designed to converge to complicated high-dimensional target distributions, to facilitate sampling. The speed of this convergence is essential for practical use. In this talk, we will present several theoretical probability results which can help improve the Metropolis algorithm's convergence speed. Specific topics will include: diffusion limits, optimal scaling, optimal proposal shape, tempering, adaptive MCMC, the Containment property, and the notion of adversarial Markov chains. The ideas will be illustrated using the simple graphical example available at probability.ca/met. No particular background knowledge will be assumed.

Other Information: 

Location: ESB 5104 Or register in advance to attend by Zoom -- contact Ed Perkins for details.


Time: 11am Pacific 


Note: A coffee reception will be served at ESB 4133 (PIMS Lounge) beginning at 10.30am.