Local antithetic sampling with scrambled nets

DOI10.1214/07-AOS548MaRDI QIDQ955143zbMATH OpenFDO

Authors Art B. Owen

Publication date 18 November 2008

Published in The Annals of Statistics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/0811.0528

zbMATH Keywords

cubatureANOVA decompositionroot mean square errorDigital netsvariance reduction methodsb-ary reflections and foldsDigital scramblesMonomial rulesmultidimensional quadraturesRandomized quasi-Monte Carlo methodVariance for scrambled digital nets

Mathematics Subject Classification ID

Monte Carlo methods (65C05) Multidimensional problems (41A63) Approximate quadratures (41A55) Pseudo-random numbers; Monte Carlo methods (11K45)

Abstract: We consider the problem of computing an approximation to the integral $I=in{t}_{[0,1{]}^d}f(x)dx$. Monte Carlo (MC) sampling typically attains a root mean squared error (RMSE) of $O(n^{-1/2})$ from $n$ independent random function evaluations. By contrast, quasi-Monte Carlo (QMC) sampling using carefully equispaced evaluation points can attain the rate $O(n^{-1+varepsilon})$ for any $varepsilon>0$ and randomized QMC (RQMC) can attain the RMSE $O(n^{-3/2+varepsilon})$, both under mild conditions on $f$. Classical variance reduction methods for MC can be adapted to QMC. Published results combining QMC with importance sampling and with control variates have found worthwhile improvements, but no change in the error rate. This paper extends the classical variance reduction method of antithetic sampling and combines it with RQMC. One such method is shown to bring a modest improvement in the RMSE rate, attaining $O(n^{-3/2-1/d+varepsilon})$ for any $varepsilon>0$, for smooth enough $f$.

Recommendations

• Barren plateaus from learning scramblers with local cost functions
• Approximation by neural networks with scattered data
• Scrambling non-uniform nets
• An Acceleration Strategy for Randomize-Then-Optimize Sampling Via Deep Neural Networks
• Neural networks-based variationally enhanced sampling
• Scattered data interpolation based upon generalized minimum norm networks
• Neural Networks for Localized Approximation
• scientific article; zbMATH DE number 7306864
• Localized sampling in the presence of noise

Cites work

• scientific article; zbMATH DE number 53679 (Why is no real title available?)
• scientific article; zbMATH DE number 3591295 (Why is no real title available?)
• scientific article; zbMATH DE number 1103058 (Why is no real title available?)
• scientific article; zbMATH DE number 1999206 (Why is no real title available?)
• scientific article; zbMATH DE number 822320 (Why is no real title available?)
• scientific article; zbMATH DE number 914714 (Why is no real title available?)
• scientific article; zbMATH DE number 3302899 (Why is no real title available?)
• A Class of Statistics with Asymptotically Normal Distribution
• Control variates for quasi-Monte Carlo (with comments and rejoinder)
• Discrépance de suites associées à un système de numération (en dimension s)
• Funktionen von beschränkter Variation in der Theorie der Gleichverteilung
• Geometric discrepancy. An illustrated guide
• MinT: A Database for Optimal Net Parameters
• Monomial cubature rules since ``Stroud: A compilation
• Monomial cubature rules since ``Stroud: A compilation. II
• Monte Carlo Variance of Scrambled Net Quadrature
• Numerical Evaluation of Multiple Integrals
• Quasi-Random Methods for Estimating Integrals Using Relatively Small Samples
• Scrambled net variance for integrals of smooth functions
• Scrambling Sobol' and Niederreiter-Xing points
• Tensor Analysis of ANOVA Decomposition
• The discrepancy and gain coefficients of scrambled digital nets.
• The jackknife estimate of variance
• The microstructure of \((t,m,s)\)-nets
• Variance with alternative scramblings of digital nets
• \(I\)-binomial scrambling of digital nets and sequences

Cited in

(21)

• On Dropping the First Sobol’ Point
• Monte Carlo with determinantal point processes
• Quasi-Monte Carlo integration using digital nets with antithetics
• Improving Approximate Bayesian Computation via Quasi-Monte Carlo
• On the Error Rate of Importance Sampling with Randomized Quasi-Monte Carlo
• Are quasi-Monte Carlo algorithms efficient for two-stage stochastic programs?
• Preintegration via Active Subspace
• scientific article; zbMATH DE number 7626758 (Why is no real title available?)
• On integration methods based on scrambled nets of arbitrary size
• Quasi-Monte Carlo methods for lattice systems: a first look
• Conditional quasi-Monte Carlo with constrained active subspaces
• Decision-theoretic sensitivity analysis for reservoir development under uncertainty using multilevel quasi-Monte Carlo methods
• Scrambling non-uniform nets
• Achieving high convergence rates by quasi-Monte Carlo and importance sampling for unbounded integrands
• A Strong Law of Large Numbers for Scrambled Net Integration
• Higher order scrambled digital nets achieve the optimal rate of the root mean square error for smooth integrands
• Quasi-Random Sampling for Multivariate Distributions via Generative Neural Networks
• Van der Corput and Golden ratio sequences along the Hilbert space-filling curve
• Monte Carlo integration of non-differentiable functions on \([0,1]^\iota\), \(\iota =1,\ldots, d\), using a single determinantal point pattern defined on \([0,1]^d\)
• Scrambled geometric net integration over general product spaces
• Sensitivity estimation of conditional value at risk using randomized quasi-Monte Carlo

Describes a project that uses

Uses Software

• MinT

This page was built for publication: Local antithetic sampling with scrambled nets

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q955143)