A note on the sample complexity of multi-target detection

TL;DR

This paper analyzes the sample complexity of multi-target detection in noisy environments, providing bounds and algorithms relevant to cryo-EM and related fields.

Contribution

It establishes fundamental upper and lower bounds for multi-target detection models in high-noise settings, connecting to multi-reference alignment and autocorrelation methods.

Findings

Lower bounds via reduction to multi-reference alignment

Upper bounds using autocorrelation-based recovery algorithms

Insights into estimation limits in noisy, multi-target scenarios

Abstract

This work studies the sample complexity of the multi-target detection (MTD) problem, which involves recovering a signal from a noisy measurement containing multiple instances of a target signal in unknown locations, each transformed by a random group element. This problem is primarily motivated by single-particle cryo-electron microscopy (cryo-EM), a groundbreaking technology for determining the structures of biological molecules. We establish upper and lower bounds for various MTD models in the high-noise regime as a function of the group, the distribution over the group, and the arrangement of signal occurrences within the measurement. The lower bounds are established through a reduction to the related multi-reference alignment problem, while the upper bounds are derived from explicit recovery algorithms utilizing autocorrelation analysis. These findings provide fundamental insights into estimation limits in noisy environments and lay the groundwork for extending this analysis to more complex applications, such as cryo-EM.