Conference Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 268 - SBSS Student Paper Competition II
Type: Topic Contributed
Date/Time: Tuesday, August 9, 2022 : 10:30 AM to 12:20 PM
Sponsor: Section on Bayesian Statistical Science
Abstract #322175
Title: Vector-Valued Control Variates
Author(s): Zhuo Sun* and Alessandro Barp and Fran├žois-Xavier Briol
Companies: University College London and University of Cambridge and University College London
Keywords: Control Variates; RKHS; Stein's method; Multi-task Learning; Markov Chain Monte Carlo; Variance Reduction

Control variates are post-processing tools for Monte Carlo estimators which can lead to significant variance reduction. This approach usually requires a large number of samples, which can be prohibitive for applications where sampling from a posterior or evaluating the integrand is computationally expensive. Furthermore, there are many scenarios where we need to compute multiple related integrals simultaneously or sequentially, which can further exacerbate computational costs. In this paper, we propose vector-valued control variates, an extension of control variates which can be used to reduce the variance of multiple integrals jointly. This allows the transfer of information across integration tasks, and hence reduces the overall requirement for a large number of samples. We focus on control variates based on kernel interpolants and our novel construction is obtained through a generalised Stein identity and the development of novel matrix-valued Stein reproducing kernels. We demonstrate our methodology on a range of problems including multifidelity modelling and model evidence computation through thermodynamic integration.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2022 program