Texto completo
|
PDF (Document Portable Format)
- Se necesita un visor de ficheros PDF, como GSview, Xpdf o Adobe Acrobat Reader
Descargar (901kB) | Vista Previa |
Valcarcel Macua, Sergio; Zazo Bello, Santiago y Zazo Muncharaz, Javier (2015). Distributed black-box optimization of nonconvex functions. En: "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.
Título: | Distributed black-box optimization of nonconvex functions |
---|---|
Autor/es: |
|
Tipo de Documento: | Ponencia en Congreso o Jornada (Artículo) |
Título del Evento: | IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) |
Fechas del Evento: | 19/04/2015 - 24/04/2015 |
Lugar del Evento: | Brisbane, QLD, Australia |
Título del Libro: | IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) |
Fecha: | Abril 2015 |
Materias: | |
Palabras Clave Informales: | Adaptive networks, cross-entropy, diffusion strategies, global optimization, stochastic approximation |
Escuela: | E.T.S.I. Telecomunicación (UPM) |
Departamento: | Señales, Sistemas y Radiocomunicaciones |
Licencias Creative Commons: | Reconocimiento - Sin obra derivada - No comercial |
|
PDF (Document Portable Format)
- Se necesita un visor de ficheros PDF, como GSview, Xpdf o Adobe Acrobat Reader
Descargar (901kB) | Vista Previa |
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.
Tipo | Código | Acrónimo | Responsable | Título |
---|---|---|---|---|
Gobierno de España | TEC2013-46011-C3-1-R | UnderWorld | Sin especificar | Sin especificar |
Gobierno de España | CSD2008-00010 COMONSENS | CONSOLIDER-INGENIO 2010 | Sin especificar | Sin especificar |
ID de Registro: | 46916 |
---|---|
Identificador DC: | http://oa.upm.es/46916/ |
Identificador OAI: | oai:oa.upm.es:46916 |
Identificador DOI: | 10.1109/ICASSP.2015.7178640 |
URL Oficial: | http://ieeexplore.ieee.org/document/7178640/ |
Depositado por: | Memoria Investigacion |
Depositado el: | 21 Jun 2017 15:39 |
Ultima Modificación: | 21 Jun 2017 15:39 |