On Large-Scale Multiple Testing Over Networks: An Asymptotic Approach

TL;DR

This paper introduces two asymptotic methods, proportion-matching and greedy aggregation, for efficient large-scale multiple testing over networks, achieving near-optimal power with minimal communication.

Contribution

It develops novel asymptotic approaches tailored for distributed network testing, providing explicit optimality characterizations and convergence rates.

Findings

Proportion-matching achieves global BH performance with minimal communication.

Greedy aggregation effectively approximates optimal rejection regions.

Both methods have proven convergence rates for FDR and power.

Abstract

This work concerns developing communication- and computation-efficient methods for large-scale multiple testing over networks, which is of interest to many practical applications. We take an asymptotic approach and propose two methods, proportion-matching and greedy aggregation, tailored to distributed settings. The proportion-matching method achieves the global BH performance yet only requires a one-shot communication of the (estimated) proportion of true null hypotheses as well as the number of p-values at each node. By focusing on the asymptotic optimal power, we go beyond the BH procedure by providing an explicit characterization of the asymptotic optimal solution. This leads to the greedy aggregation method that effectively approximates the optimal rejection regions at each node, while computation efficiency comes from the greedy-type approach naturally. Moreover, for both methods, we provide the rate of convergence for both the FDR and power. Extensive numerical results over a variety of challenging settings are provided to support our theoretical findings.