We use Array-RQMC sampling in a walk on spheres (WoS) algorithm for Dirichlet boundary value problems. On a collection of problems, we find that Array-RQMC-WoS reduces the Monte Carlo MSE or variance by factors ranging from $71$-fold to $3087$-fold at $n=2^{17}$ trajectories. The variance is known to be $o(1/n)$ but attains empirical rates between $n^{-1.4}$ and $n^{-1.8}$ in our examples. A simpler RQMC-WoS algorithm studied in Ho and Owen (2026) has more theoretical support but only reduced variance by 1.8 to 10.7-fold on the same set of examples. In order to explain this improvement, we introduce a column-wise mean dimension of the RQMC error based on Sobol' indices. It matches the usual mean dimension for Monte Carlo and the mean dimension of a dual lattice error for randomized lattices. We find for a gasket example from Crane et al. (2025) that the mean dimension of Array-RQMC-WoS errors is much higher than an analogous Array-MC-WoS algorithm has. v2 replaced v1's QMCPy lattice with Korobov lattices from LatNet Builder, but left the old abstract in the meta-data v3 corrects the v2 abstract in this meta data
We use Array-RQMC sampling in a walk on spheres (WoS) algorithm for Dirichlet boundary value problems. On a collection of problems, we find that Array-RQMC-WoS reduces the Monte Carlo MSE or variance by factors ranging from 71-fold to 3087-fold at n=217 trajectories. The variance is known to be o(1/n) but attains empirical rates between n−1.4 and n−1.8 in our examples. A simpler RQMC-WoS algorithm studied in Ho and Owen (2026) has more theoretical support but only reduced variance by 1.8 to 10.7-fold on the same set of examples. In order to explain this improvement, we introduce a column-wise mean dimension of the RQMC error based on Sobol' indices. It matches the usual mean dimension for Monte Carlo and the mean dimension of a dual lattice error for randomized lattices. We find for a gasket example from Crane et al. (2025) that the mean dimension of Array-RQMC-WoS errors is much higher than an analogous Array-MC-WoS algorithm has. v2 replaced v1's QMCPy lattice with Korobov lattices from LatNet Builder, but left the old abstract in the meta-data v3 corrects the v2 abstract in this meta data
This paper studies randomised quasi-Monte Carlo (RQMC) methods for Metropolis--Hastings (MH) algorithms, whose accept--reject discontinuity complicates RQMC analysis. To circumvent some of these complications, we study Array-RQMC for MH, with N replicated chains evolved over a fixed time horizon T, where we make use of Hilbert sorting for joint multidimensional updates at each step. For one-dimensional MH and bounded-variation test functions, we prove a sampling mean squared error bound of OT(N−4/3) relative to the finite-horizon expectation. For product targets, coordinatewise Array-RQMC Metropolis-within-Gibbs retains this exponent for additive marginal bounded-variation tests, with constants that may depend on dimension. With Gaussian random-walk proposals, the bounds extend to scalar and additive marginal polynomials with an additional factor (logN)ℓ, where ℓ is the maximum polynomial degree. For joint updates in state dimension p using p proposal uniforms, we provide a conditional OT(N−1−1/(p+2)) framework under regularity assumptions that are not verified for the joint multidimensional examples. Numerical experiments support the verified one-dimensional theory and demonstrate variance reduction for joint Hilbert-sorted updates, including Bayesian logistic regression, while illustrating the limitations of these gains as dimension increases.
Abstract Randomized quasi-Monte Carlo (RQMC) methods estimate the mean of a random variable by sampling an integrand at n equidistributed points. For scrambled digital nets, the resulting variance is typically O~(n−θ), where θ∈[1,3] depends on the smoothness of the integrand and O~ neglects logarithmic factors. While RQMC can be far more accurate than plain Monte Carlo (MC), it remains difficult to get confidence intervals on RQMC estimates. We investigate some empirical Bernstein confidence intervals (EBCIs) and hedged betting confidence intervals (HBCIs), both from Waudby-Smith and Ramdas (2024, J. Roy. Statist. Soc. B, 86, 1–27), when the random variable of interest is subject to known bounds. When there are N integrand evaluations partitioned into R independent replicates of n=N/R RQMC points, and the RQMC variance is Θ(n−θ), then an oracle minimizing the width of a Bennett confidence interval would choose n=Θ(N1/(θ+1)). The resulting intervals have a width Θ(N−θ/(θ+1)). Our empirical investigations had optimal values of n grow slowly with N, HBCI intervals that were usually narrower than the EBCI ones and optimal values of n for HBCI that were equal to or smaller than the ones for the oracle.
This paper studies the convergence rate of randomized quasi--Monte Carlo (RQMC) for discontinuous functions, which are often of infinite variation in the sense of Hardy and Krause. It was previously known that the root mean square error (RMSE) of RQMC is only o(n−1/2) for discontinuous functions. For certain discontinuous functions in d dimensions, we prove that the RMSE of RQMC is O(n−1/2−1/(4d−2)+ϵ) for any ϵ>0 and arbitrary n. If some discontinuity boundaries are parallel to some coordinate axes, the rate can be improved to O(n−1/2−1/(4du−2)+ϵ), where du denotes the so-called irregular dimension, that is, the number of axes which are not parallel to the discontinuity boundaries. Moreover, this paper shows that the RMSE is O(n−1/2−1/(2d)) for certain indicator functions.
Abstract A crude Monte Carlo (MC) method allows to calculate integrals over a d-dimensional cube. As the number N of integration nodes becomes large, the rate of probable error of the MC method decreases as O ( 1 / N ) {O(1/\sqrt{N})} . The use of quasi-random points instead of random points in the MC algorithm converts it to the quasi-Monte Carlo (QMC) method. The asymptotic error estimate of QMC integration of d-dimensional functions contains a multiplier 1 / N {1/N} . However, the multiplier ( ln N ) d {(\ln N)^{d}} is also a part of the error estimate, which makes it virtually useless. We have proved that, in the general case, the QMC error estimate is not limited to the factor 1 / N {1/N} . However, our numerical experiments show that using quasi-random points of Sobol sequences with N = 2 m {N=2^{m}} with natural m makes the integration error approximately proportional to 1 / N {1/N} . In our numerical experiments, d ≤ 15 {d\leq 15} , and we used N ≤ 2 40 {N\leq 2^{40}} points generated by the SOBOLSEQ16384 code published in 2011. In this code, d ≤ 2 14 {d\leq 2^{14}} and N ≤ 2 63 {N\leq 2^{63}} .
This study investigates the performance of median-of-means sampling compared to traditional mean-of-means sampling for computing the Keister function integral using Randomized Quasi-Monte Carlo (RQMC) methods. The research tests both lattice points and digital nets as point distributions across dimensions 2, 3, 5, and 8, with sample sizes ranging from 2^8 to 2^19 points. Results demonstrate that median-of-means sampling consistently outperforms mean-of-means for sample sizes larger than 10^3 points, while mean-of-means shows better accuracy with smaller sample sizes, particularly for digital nets. The study also confirms previous theoretical predictions about median-of-means' superior performance with larger sample sizes and reflects the known challenges of maintaining accuracy in higher-dimensional integration. These findings support recent research suggesting median-of-means as a promising alternative to traditional sampling methods in numerical integration, though limitations in sample size and dimensionality warrant further investigation with different test functions and larger parameter spaces.
We study the numerical approximation of high-dimensional integrals using randomized quasi-Monte Carlo (RQMC) methods, with a focus on scrambled Sobol' sequences. While asymptotically faster than Monte Carlo, classical RQMC suffers from error bounds that grow exponentially in the dimension s, and its root mean squared error (RMSE) convergence rate is generally no better than O(N−3/2). To overcome these limitations, we combine two recent developments: Suzuki's coarse scrambling and the median trick. Coarse scrambling randomizes Sobol' sequences according to their generating base polynomials and significantly reduces the maximal gain coefficient. By appropriately choosing the base polynomials, we show that the coefficient can be made uniformly bounded in s, yielding an O(N−1/2) RMSE for L2 integrands with a dimension-independent constant. The median trick then enables near-optimal convergence for function classes beyond L2: for Lp integrands with p∈(1,2), the median of independent coarsely scrambled estimates achieves an error of O(N−1+1/p) with high probability; for integrands in the Haar wavelet space Hwav,α,s,p,q with αp:=α−(1/p−1/2)+>0, we prove a high-probability error bound of O(N−αp−1/2+ε) for any ε>0; the bound improves to O(N−r−αp−1/2+ε) when the integrand has dominating mixed derivatives of order r≥1 that belong to Hwav,α,s,p,q. The latter two bounds are uniform in s under suitable conditions on the ANOVA components of the integrands. Numerical experiments confirm the predicted convergence rates and demonstrate the robustness of the proposed approach.