Distributed black-box optimization of nonconvex functions

Valcarcel Macua, Sergio and Zazo Bello, Santiago and Zazo Muncharaz, Javier (2015). Distributed black-box optimization of nonconvex functions. In: "IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)", 19/04/2015 - 24/04/2015, Brisbane, QLD, Australia. pp. 3591-3595. https://doi.org/10.1109/ICASSP.2015.7178640.

Description

Title: Distributed black-box optimization of nonconvex functions
Author/s:
  • Valcarcel Macua, Sergio
  • Zazo Bello, Santiago
  • Zazo Muncharaz, Javier
Item Type: Presentation at Congress or Conference (Article)
Event Title: IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
Event Dates: 19/04/2015 - 24/04/2015
Event Location: Brisbane, QLD, Australia
Title of Book: IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
Date: April 2015
Subjects:
Freetext Keywords: Adaptive networks, cross-entropy, diffusion strategies, global optimization, stochastic approximation
Faculty: E.T.S.I. Telecomunicación (UPM)
Department: Señales, Sistemas y Radiocomunicaciones
Creative Commons Licenses: Recognition - No derivative works - Non commercial

Full text

[img]
Preview
PDF - Requires a PDF viewer, such as GSview, Xpdf or Adobe Acrobat Reader
Download (901kB) | Preview

Abstract

We combine model-based methods and distributed stochastic approximation to propose a fully distributed algorithm for nonconvex optimization, with good empirical performance and convergence guarantees. Neither the expression of the objective nor its gradient are known. Instead, the objective is like a “black-box”, in which the agents input candidate solutions and evaluate the output. Without central coordination, the distributed algorithm naturally balances the computational load among the agents. This is especially relevant when many samples are needed (e.g., for high-dimensional objectives) or when evaluating each sample is costly. Numerical experiments over a difficult benchmark show that the networked agents match the performance of a centralized architecture, being able to approach the global optimum, while none of the individual noncooperative agents could by itself.

Funding Projects

TypeCodeAcronymLeaderTitle
Government of SpainTEC2013-46011-C3-1-RUnderWorldUnspecifiedUnspecified
Government of SpainCSD2008-00010 COMONSENSCONSOLIDER-INGENIO 2010UnspecifiedUnspecified

More information

Item ID: 46916
DC Identifier: http://oa.upm.es/46916/
OAI Identifier: oai:oa.upm.es:46916
DOI: 10.1109/ICASSP.2015.7178640
Official URL: http://ieeexplore.ieee.org/document/7178640/
Deposited by: Memoria Investigacion
Deposited on: 21 Jun 2017 15:39
Last Modified: 21 Jun 2017 15:39
  • Logo InvestigaM (UPM)
  • Logo GEOUP4
  • Logo Open Access
  • Open Access
  • Logo Sherpa/Romeo
    Check whether the anglo-saxon journal in which you have published an article allows you to also publish it under open access.
  • Logo Dulcinea
    Check whether the spanish journal in which you have published an article allows you to also publish it under open access.
  • Logo de Recolecta
  • Logo del Observatorio I+D+i UPM
  • Logo de OpenCourseWare UPM