CCM Colloquium: Nawaf Bou-Rabee (Rutgers University)

Event

When

  • – ET

  • All Day

  • All Day

  • All Day

  • All Day

  • All Day

  • All Day

  • All Day

  • All Day

  • All Day

  • All Day

  • All Day

  • All Day

  • – ET

Where

162 Fifth Avenue, 3rd Floor Classroom

160 Fifth Avenue

NY, United States

Title: Adjustable Randomized Time Integrators for Hamiltonian Monte Carlo

 

Abstract: Hamiltonian Monte Carlo is a gradient-based Metropolis-Hastings algorithm with a proposal distribution that involves ceil(T/h) steps of a time integrator of Hamiltonian dynamics; where T and h are the duration and time-step-size hyperparameters.  Here we show how to incorporate time integrator randomization into the proposal distribution and present L^2 Wasserstein mixing time guarantees for the corresponding unadjusted algorithm.  These guarantees illustrate the potential role of time integrator randomization in HMC.
 

More Events

October 15–16, 2026

October 15, 2026

1–5 p.m. ET