Multireference Alignment is Easier with an Aperiodic Translation Distribution

TL;DR

This paper demonstrates that multireference alignment becomes easier with an aperiodic translation distribution, reducing the sample complexity from (1/SNR^3) to (1/SNR^2) in low SNR regimes, using spectral and optimization algorithms.

Contribution

It generalizes the sample complexity analysis for multireference alignment to aperiodic distributions, showing improved rates and proposing new algorithms.

Findings

Sample complexity scales as (1/SNR^2) for aperiodic distributions.

A simple spectral algorithm achieves the optimal rate.

Connections are established between multireference alignment and the spiked covariance model.

Abstract

In the multireference alignment model, a signal is observed by the action of a random circular translation and the addition of Gaussian noise. The goal is to recover the signal's orbit by accessing multiple independent observations. Of particular interest is the sample complexity, i.e., the number of observations/samples needed in terms of the signal-to-noise ratio (the signal energy divided by the noise variance) in order to drive the mean-square error (MSE) to zero. Previous work showed that if the translations are drawn from the uniform distribution, then, in the low SNR regime, the sample complexity of the problem scales as $\omega(1/\text{SNR}^3)$. In this work, using a generalization of the Chapman--Robbins bound for orbits and expansions of the $\chi^2$ divergence at low SNR, we show that in the same regime the sample complexity for any aperiodic translation distribution scales as $\omega(1/\text{SNR}^2)$. This rate is achieved by a simple spectral algorithm. We propose two additional algorithms based on non-convex optimization and expectation-maximization. We also draw a connection between the multireference alignment problem and the spiked covariance model.