Variance Reduction in Stochastic Reaction Networks using Control Variates

TL;DR

This paper introduces a control variate method based on statistical moments to reduce variance in Monte Carlo estimations for stochastic reaction networks, improving efficiency through an optimized subset selection algorithm.

Contribution

It proposes a novel variance reduction technique using moment-based control variates and an efficient subset selection algorithm for stochastic reaction networks.

Findings

Significant variance reduction demonstrated in case studies

Efficient subset selection improves computational performance

Method outperforms traditional Monte Carlo estimators

Abstract

Monte Carlo estimation in plays a crucial role in stochastic reaction networks. However, reducing the statistical uncertainty of the corresponding estimators requires sampling a large number of trajectories. We propose control variates based on the statistical moments of the process to reduce the estimators' variances. We develop an algorithm that selects an efficient subset of infinitely many control variates. To this end, the algorithm uses resampling and a redundancy-aware greedy selection. We demonstrate the efficiency of our approach in several case studies.