DeepAI AI Chat
Log In Sign Up

Decentralized Inexact Proximal Gradient Method With Network-Independent Stepsizes for Convex Composite Optimization

by   Luyao Guo, et al.

This paper considers decentralized convex composite optimization over undirected and connected networks, where the local loss function contains both smooth and nonsmooth terms. For this problem, a novel CTA (Combine-Then-Adapt)-based decentralized algorithm is proposed under uncoordinated network-independent constant stepsizes. Particularly, the proposed algorithm only needs to approximately solve a sequence of proximal mappings, which benefits the decentralized composite optimization where the proximal mappings of the nonsmooth loss functions may not have analytic solutions. For the general convex case, we prove the O(1/k) convergence rate of the proposed algorithm, which can be improved to o(1/k) if the proximal mappings are solved exactly. Moreover, with metric subregularity, we establish the linear convergence rate. Finally, the numerical experiments demonstrate the efficiency of the algorithm.


A decentralized proximal-gradient method with network independent step-sizes and separated convergence rates

This paper considers the problem of decentralized optimization with a co...

A Linearly Convergent Proximal Gradient Algorithm for Decentralized Optimization

Decentralized optimization is a promising paradigm that finds various ap...

Decentralized Distributed Optimization for Saddle Point Problems

We consider distributed convex-concave saddle point problems over arbitr...

Composite Optimization Algorithms for Sigmoid Networks

In this paper, we use composite optimization algorithms to solve sigmoid...

Fast and Robust Sparsity Learning over Networks: A Decentralized Surrogate Median Regression Approach

Decentralized sparsity learning has attracted a significant amount of at...

Decentralized Composite Optimization in Stochastic Networks: A Dual Averaging Approach with Linear Convergence

Decentralized optimization, particularly the class of decentralized comp...

A family of multi-parameterized proximal point algorithms

In this paper, a multi-parameterized proximal point algorithm combining ...