error
Kurzer Serviceunterbruch am Donnerstag, 22. Januar 2026, 12 bis 13 Uhr. Sie können in diesem Zeitraum keine neuen Dokumente hochladen oder bestehende Einträge bearbeiten. Das Login wird in diesem Zeitraum deaktiviert. Grund: Wartungsarbeiten // Short service interruption on Thursday, January 22, 2026, 12.00 – 13.00. During this time, you won’t be able to upload new documents or edit existing records. The login will be deactivated during this time. Reason: maintenance work
 

Nonconvex landscapes for Z2 synchronization and graph clustering are benign near exact recovery thresholds


METADATA ONLY
Loading...

Date

2024-07-18

Publication Type

Working Paper

ETH Bibliography

yes

Citations

Altmetric
METADATA ONLY

Data

Rights / License

Abstract

We study the optimization landscape of a smooth nonconvex program arising from synchronization over the two-element group $\mathbf{Z}_2$, that is, recovering $z_1, \dots, z_n \in \{\pm 1\}$ from (noisy) relative measurements $R_{ij} \approx z_i z_j$. Starting from a max-cut--like combinatorial problem, for integer parameter $r \geq 2$, the nonconvex problem we study can be viewed both as a rank-$r$ Burer--Monteiro factorization of the standard max-cut semidefinite relaxation and as a relaxation of $\{ \pm 1 \}$ to the unit sphere in $\mathbf{R}^r$. First, we present deterministic, non-asymptotic conditions on the measurement graph and noise under which every second-order critical point of the nonconvex problem yields exact recovery of the ground truth. Then, via probabilistic analysis, we obtain asymptotic guarantees for three benchmark problems: (1) synchronization with a complete graph and Gaussian noise, (2) synchronization with an Erdős--Rényi random graph and Bernoulli noise, and (3) graph clustering under the binary symmetric stochastic block model. In each case, we have, asymptotically as the problem size goes to infinity, a benign nonconvex landscape near a previously-established optimal threshold for exact recovery; we can approach this threshold to arbitrary precision with large enough (but finite) rank parameter $r$. In addition, our results are robust to monotone adversaries.

Publication status

published

Editor

Book title

Journal / series

Volume

Pages / Article No.

2407.13407

Publisher

Cornell University

Event

Edition / version

v1

Methods

Software

Geographic location

Date collected

Date created

Subject

Optimization and Control (math.OC); Statistics Theory (math.ST); FOS: Mathematics

Organisational unit

09679 - Bandeira, Afonso / Bandeira, Afonso check_circle

Notes

Funding

Related publications and datasets

Is previous version of: