A Universal Decomposition for Distributed Optimization Algorithms

arXiv2206.07096MaRDI QIDQ6402115FDOQ6402115

Publication date: 14 June 2022

Abstract: In the distributed optimization problem for a multi-agent system, each agent knows a local function and must find a minimizer of the sum of all agents' local functions by performing a combination of local gradient evaluations and communicating information with neighboring agents. We prove that every distributed optimization algorithm can be factored into a centralized optimization method and a second-order consensus estimator, effectively separating the "optimization" and "consensus" tasks. We illustrate this fact by providing the decomposition for many recently proposed distributed optimization algorithms. Conversely, we prove that any optimization method that converges in the centralized setting can be combined with any second-order consensus estimator to form a distributed optimization algorithm that converges in the multi-agent setting. Finally, we describe how our decomposition may lead to a more systematic algorithm design methodology.

This page was built for publication: A Universal Decomposition for Distributed Optimization Algorithms

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6402115)