DeepAI AI Chat
Log In Sign Up

Measuring Sample Quality with Stein's Method

by   Jackson Gorham, et al.

To improve the efficiency of Monte Carlo estimation, practitioners are turning to biased Markov chain Monte Carlo procedures that trade off asymptotic exactness for computational speed. The reasoning is sound: a reduction in variance due to more rapid sampling can outweigh the bias introduced. However, the inexactness creates new challenges for sampler and parameter selection, since standard measures of sample quality like effective sample size do not account for asymptotic bias. To address these challenges, we introduce a new computable quality measure based on Stein's method that quantifies the maximum discrepancy between sample and target expectations over a large class of test functions. We use our tool to compare exact, biased, and deterministic sample sequences and illustrate applications to hyperparameter selection, convergence rate assessment, and quantifying bias-variance tradeoffs in posterior inference.


page 1

page 2

page 3

page 4


Measuring Sample Quality with Diffusions

Standard Markov chain Monte Carlo diagnostics, like effective sample siz...

Measuring Sample Quality with Kernels

Approximate Markov chain Monte Carlo (MCMC) offers the promise of more r...

Globally-centered autocovariances in MCMC

Autocovariances are a fundamental quantity of interest in Markov chain M...

Bias-Variance Trade-off and Overlearning in Dynamic Decision Problems

Modern Monte Carlo-type approaches to dynamic decision problems face the...

Efficient estimation of autocorrelation spectra

The performance of Markov chain Monte Carlo calculations is determined b...

State and parameter learning with PaRIS particle Gibbs

Non-linear state-space models, also known as general hidden Markov model...

Code Repositories


Code to compute the Stein discrepancy between a sample distribution and its target

view repo