Provably Accelerated Decentralized Gradient Method Over Unbalanced Directed Graphs

07/26/2021
by   Zhuoqing Song, et al.
0

In this work, we consider the decentralized optimization problem in which a network of n agents, each possessing a smooth and convex objective function, wish to collaboratively minimize the average of all the objective functions through peer-to-peer communication in a directed graph. To solve the problem, we propose two accelerated Push-DIGing methods termed APD and APD-SC for minimizing non-strongly convex objective functions and strongly convex ones, respectively. We show that APD and APD-SC respectively converge at the rates O(1/k^2) and O((1 - C√(μ/L))^k) up to constant factors depending only on the mixing matrix. To the best of our knowledge, APD and APD-SC are the first decentralized methods to achieve provable acceleration over unbalanced directed graphs. Numerical experiments demonstrate the effectiveness of both methods.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
03/20/2018

A Push-Pull Gradient Method for Distributed Optimization in Networks

In this paper, we focus on solving a distributed convex optimization pro...
research
12/03/2021

A Divide-and-Conquer Algorithm for Distributed Optimization on Networks

In this paper, we consider networks with topologies described by some co...
research
01/14/2023

CEDAS: A Compressed Decentralized Stochastic Gradient Method with Improved Convergence

In this paper, we consider solving the distributed optimization problem ...
research
10/15/2018

Push-Pull Gradient Methods for Distributed Optimization in Networks

In this paper, we focus on solving a distributed convex optimization pro...
research
06/14/2021

Compressed Gradient Tracking for Decentralized Optimization Over General Directed Networks

In this paper, we propose two communication-efficient algorithms for dec...
research
06/08/2022

Push–Pull with Device Sampling

We consider decentralized optimization problems in which a number of age...
research
06/21/2023

Distributed Random Reshuffling Methods with Improved Convergence

This paper proposes two distributed random reshuffling methods, namely G...

Please sign up or login with your details

Forgot password? Click here to reset