Searcharxiv⌕ Search

arXiv subjects

Leah F. South

Publications and source records attributed to Leah F. South.

8 recordsLinked to original sources
arXiv2026

$λ$-PSD: Scalable Approximate SNR-Optimised Polynomial Stein Discrepancies

Polynomial Stein discrepancies (PSD) provide a scalable alternative to kernel Stein methods for measuring sample quality and goodness-of-fit testing, but their statistical properties remain poorly understood. We show that increasing polynomial degree primarily amplifies signal without adequately controlling variance, rather than directly optimising the signal-to-noise ratio (SNR). Under suitable assumptions, this might lead to a failure mode in which the $\text{SNR}^2$ can provably decay exponentially with polynomial degree. Motivated by this observation, we reformulate Stein discrepancy construction as an explicit $\text{SNR}^2$ maximisation problem, yielding a Rayleigh quotient over Stein features. This perspective motivates $λ$-PSD, an approximate scalable covariance-aware reweighting scheme defined in a low-dimensional subspace. Under Gaussian settings, we show that $λ$-PSD avoids the exponential $\text{SNR}^2$ collapse and achieves a stable $\text{SNR}^2$. Empirically, $λ$-PSD substantially improves test power while retaining linear-time complexity in the number of samples, highlighting the importance of SNR-aware design for scalable Stein discrepancies.

stat.ME↗
arXiv2025

Dynamic Learning Rate for Deep Reinforcement Learning: A Bandit Approach

In deep Reinforcement Learning (RL), the learning rate critically influences both stability and performance, yet its optimal value shifts during training as the environment and policy evolve. Standard decay schedulers assume monotonic convergence and often misalign with these dynamics, leading to premature or delayed adjustments. We introduce LRRL, a meta-learning approach that dynamically selects the learning rate based on policy performance rather than training steps. LRRL adaptively favors rates that improve returns, remaining robust even when the candidate set includes values that individually cause divergence. Across Atari and MuJoCo benchmarks, LRRL achieves performance competitive with or superior to tuned baselines and standard schedulers. Our findings position LRRL as a practical solution for adapting to non-stationary objectives in deep RL.

cs.LG↗
arXiv2025

Ensemble Control Variates

Control variates have become an increasingly popular variance-reduction technique in Bayesian inference. Many broadly applicable control variates are based on the Langevin-Stein operator, which leverages gradient information from any gradient-based sampler to produce variance-reduced estimators of expectations. These control variates typically require optimising over a function $u(θ)$ within a user-defined functional class $G$, such as the space of $Q$th-order polynomials or a reproducing kernel Hilbert space. We propose using averaging-based ensemble learning to construct Stein-based control variates. While the proposed framework is broadly applicable, we focus on ensembles constructed from zero-variance control variates (ZVCV), a popular parametric approach based on solving a linear approximation problem that can easily be over-parameterised in medium-to-high dimensional settings. A common remedy is to use regularised ZVCV via penalised regression, but these methods can be prohibitively slow. We introduce ensemble ZVCV methods based on ensembles of OLS estimators and evaluate the proposed methods against established methods in the literature in a simulation study. Our results show that ensemble ZVCV methods are competitive with regularised ZVCV methods in terms of statistical efficiency, but are substantially faster. This work opens a new direction for constructing broadly applicable control variate techniques via ensemble learning.

stat.CO↗
arXiv2025

Unbiased and Consistent Nested Sampling via Sequential Monte Carlo

We introduce a new class of sequential Monte Carlo methods which reformulates the essence of the nested sampling method of Skilling (2006) in terms of sequential Monte Carlo techniques. Two new algorithms are proposed, nested sampling via sequential Monte Carlo (NS-SMC) and adaptive nested sampling via sequential Monte Carlo (ANS-SMC). The new framework allows convergence results to be obtained in the setting when Markov chain Monte Carlo (MCMC) is used to produce new samples. An additional benefit is that marginal likelihood (normalising constant) estimates given by NS-SMC are unbiased. In contrast to NS, the analysis of our proposed algorithms does not require the (unrealistic) assumption that the simulated samples be independent. We show that a minor adjustment to our ANS-SMC algorithm recovers the original NS algorithm, which provides insights as to why NS seems to produce accurate estimates despite a typical violation of its assumptions. A numerical study is conducted where the performance of the proposed algorithms and temperature-annealed SMC is compared on challenging problems. Code for the experiments is made available online at https://github.com/LeahPrice/SMC-NS .

stat.CO↗
arXiv2022

Regularized Zero-Variance Control Variates

Zero-variance control variates (ZV-CV) are a post-processing method to reduce the variance of Monte Carlo estimators of expectations using the derivatives of the log target. Once the derivatives are available, the only additional computational effort lies in solving a linear regression problem. Significant variance reductions have been achieved with this method in low dimensional examples, but the number of covariates in the regression rapidly increases with the dimension of the target. In this paper, we present compelling empirical evidence that the use of penalized regression techniques in the selection of high-dimensional control variates provides performance gains over the classical least squares method. Another type of regularization based on using subsets of derivatives, or a priori regularization as we refer to it in this paper, is also proposed to reduce computational and storage requirements. Several examples showing the utility and limitations of regularized ZV-CV for Bayesian inference are given. The methods proposed in this paper are accessible through the R package ZVCV.

stat.CO↗
arXiv2021

Post-Processing of MCMC

Markov chain Monte Carlo (MCMC) is the engine of modern Bayesian statistics, being used to approximate the posterior and derived quantities of interest. Despite this, the issue of how the output from a Markov chain is post-processed and reported is often overlooked. Convergence diagnostics can be used to control bias via burn-in removal, but these do not account for (common) situations where a limited computational budget engenders a bias-variance trade-off. The aim of this article is to review state-of-the-art techniques for post-processing Markov chain output. Our review covers methods based on discrepancy minimisation, which directly address the bias-variance trade-off, as well as general-purpose control variate methods for approximating expected quantities of interest.

stat.ME↗
arXiv2021

Semi-Exact Control Functionals From Sard's Method

The numerical approximation of posterior expected quantities of interest is considered. A novel control variate technique is proposed for post-processing of Markov chain Monte Carlo output, based both on Stein's method and an approach to numerical integration due to Sard. The resulting estimators are proven to be polynomially exact in the Gaussian context, while empirical results suggest the estimators approximate a Gaussian cubature method near the Bernstein-von-Mises limit. The main theoretical result establishes a bias-correction property in settings where the Markov chain does not leave the posterior invariant. Empirical results are presented across a selection of Bayesian inference tasks. All methods used in this paper are available in the R package ZVCV.

stat.CO↗