Decentralized conditional gradient method over time-varying graphs

TL;DR

This paper extends the distributed conditional gradient method to handle time-varying network structures, providing theoretical convergence analysis and numerical validation for dynamic graph scenarios.

Contribution

It introduces a generalized algorithm for time-varying networks and analyzes its convergence, which is a novel contribution to distributed optimization methods.

Findings

The algorithm converges under certain conditions.

Numerical experiments validate theoretical results.

Applicable to both deterministic and stochastic graph sequences.

Abstract

In this paper we study a generalization of distributed conditional gradient method to time-varying network architectures. We theoretically analyze convergence properties of the algorithm and provide numerical experiments. The time-varying network is modeled as a deterministic of a stochastic sequence of graphs.