Accelerated Dual Averaging Methods for Decentralized Constrained Optimization

TL;DR

This paper introduces two accelerated decentralized dual averaging algorithms for constrained convex optimization, achieving faster convergence rates and improved accuracy in estimating global dual variables within networked systems.

Contribution

The paper proposes two novel decentralized dual averaging algorithms with provably faster convergence and simplified communication requirements compared to existing methods.

Findings

Achieved $ ext{O}(1/t)$ convergence for general convex problems.

Developed an accelerated algorithm with $ ext{O}(1/t^2)$ convergence.

Demonstrated improved efficiency through numerical experiments.

Abstract

In this work, we study decentralized convex constrained optimization problems in networks. We focus on the dual averaging-based algorithmic framework that is well-documented to be superior in handling constraints and complex communication environments simultaneously. Two new decentralized dual averaging (DDA) algorithms are proposed. In the first one, a second-order dynamic average consensus protocol is tailored for DDA-type algorithms, which equips each agent with a provably more accurate estimate of the global dual variable than conventional schemes. We rigorously prove that the proposed algorithm attains $\mathcal{O}(1/t)$ convergence for general convex and smooth problems, for which existing DDA methods were only known to converge at $\mathcal{O}(1/\sqrt{t})$ prior to our work. In the second one, we use the extrapolation technique to accelerate the convergence of DDA. Compared to existing accelerated algorithms, where typically two different variables are exchanged among agents at each time, the proposed algorithm only seeks consensus on local gradients. Then, the extrapolation is performed based on two sequences of primal variables which are determined by the accumulations of gradients at two consecutive time instants, respectively. The algorithm is proved to converge at $\mathcal{O}(1)\left(\frac{1}{t^2}+\frac{1}{t(1-\beta)^2}\right)$, where $\beta$ denotes the second largest singular value of the mixing matrix. We remark that the condition for the algorithmic parameter to guarantee convergence does not rely on the spectrum of the mixing matrix, making itself easy to satisfy in practice. Finally, numerical results are presented to demonstrate the efficiency of the proposed methods.