Unrolled algorithms for group synchronization

TL;DR

This paper introduces unrolled algorithms for group synchronization problems, including cryo-EM, demonstrating improved performance over traditional methods through data-driven optimization.

Contribution

It applies the concept of algorithm unrolling to group synchronization, enhancing estimation accuracy in problems like cryo-EM and multi-reference alignment.

Findings

Unrolled algorithms outperform traditional methods in various scenarios.

Data-driven training improves synchronization accuracy.

Applicable to groups like 3-D rotations and multi-reference alignment.

Abstract

The group synchronization problem involves estimating a collection of group elements from noisy measurements of their pairwise ratios. This task is a key component in many computational problems, including the molecular reconstruction problem in single-particle cryo-electron microscopy (cryo-EM). The standard methods to estimate the group elements are based on iteratively applying linear and non-linear operators, and are not necessarily optimal. Motivated by the structural similarity to deep neural networks, we adopt the concept of algorithm unrolling, where training data is used to optimize the algorithm. We design unrolled algorithms for several group synchronization instances, including synchronization over the group of 3-D rotations: the synchronization problem in cryo-EM. We also apply a similar approach to the multi-reference alignment problem. We show by numerical experiments that the unrolling strategy outperforms existing synchronization algorithms in a wide variety of scenarios.