Distributed Nonconvex Optimization: Gradient-free Iterations and $ε$-Globally Optimal Solution

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: He, Zhiyu, He, Jianping, Chen, Cailian, Guan, Xinping
Format: Preprint
Published: 2020
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914783207358464
author He, Zhiyu
He, Jianping
Chen, Cailian
Guan, Xinping
author_facet He, Zhiyu
He, Jianping
Chen, Cailian
Guan, Xinping
contents Distributed optimization utilizes local computation and communication to realize a global aim of optimizing the sum of local objective functions. This article addresses a class of constrained distributed nonconvex optimization problems involving univariate objectives, aiming to achieve global optimization without requiring local evaluations of gradients at every iteration. We propose a novel algorithm named CPCA, exploiting the notion of combining Chebyshev polynomial approximation, average consensus, and polynomial optimization. The proposed algorithm is i) able to obtain $ε$-globally optimal solutions for any arbitrarily small given accuracy $ε$, ii) efficient in both zeroth-order queries (i.e., evaluations of function values) and inter-agent communication, and iii) distributed terminable when the specified precision requirement is met. The key insight is to use polynomial approximations to substitute for general local objectives, distribute these approximations via average consensus, and solve an easier approximate version of the original problem. Due to the nice analytic properties of polynomials, this approximation not only facilitates efficient global optimization, but also allows the design of gradient-free iterations to reduce cumulative costs of queries and achieve geometric convergence for solving nonconvex problems. We provide a comprehensive analysis of the accuracy and complexities of the proposed algorithm.
format Preprint
id arxiv_https___arxiv_org_abs_2008_00252
institution arXiv
publishDate 2020
record_format arxiv
spellingShingle Distributed Nonconvex Optimization: Gradient-free Iterations and $ε$-Globally Optimal Solution
He, Zhiyu
He, Jianping
Chen, Cailian
Guan, Xinping
Optimization and Control
Systems and Control
Distributed optimization utilizes local computation and communication to realize a global aim of optimizing the sum of local objective functions. This article addresses a class of constrained distributed nonconvex optimization problems involving univariate objectives, aiming to achieve global optimization without requiring local evaluations of gradients at every iteration. We propose a novel algorithm named CPCA, exploiting the notion of combining Chebyshev polynomial approximation, average consensus, and polynomial optimization. The proposed algorithm is i) able to obtain $ε$-globally optimal solutions for any arbitrarily small given accuracy $ε$, ii) efficient in both zeroth-order queries (i.e., evaluations of function values) and inter-agent communication, and iii) distributed terminable when the specified precision requirement is met. The key insight is to use polynomial approximations to substitute for general local objectives, distribute these approximations via average consensus, and solve an easier approximate version of the original problem. Due to the nice analytic properties of polynomials, this approximation not only facilitates efficient global optimization, but also allows the design of gradient-free iterations to reduce cumulative costs of queries and achieve geometric convergence for solving nonconvex problems. We provide a comprehensive analysis of the accuracy and complexities of the proposed algorithm.
title Distributed Nonconvex Optimization: Gradient-free Iterations and $ε$-Globally Optimal Solution
topic Optimization and Control
Systems and Control
url https://arxiv.org/abs/2008.00252