Le reti neurali RBF: legami e differenze con i metodi kernel.

Diego Attilio Mancuso

Le reti neurali RBF: legami e differenze con i metodi kernel.

Translated title of the contribution: [Autom. eng. transl.] RBF neural networks: links and differences with kernel methods.

Diego Attilio Mancuso

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

The Radial Basis Function (RBF) neural networks are nonparametric regression tools similar in formula to the Nadaraya-Watson kernel estimator. Starting from this resemblance, the relationships between the RBF neural nets and the kernel methods are discussed. Some results regarding the moment matrix of the radial basis functions are used to understand the linkage between these estimators and to show a different behaviour of their bandwidth. It is also recalled that the rate of convergence of RBF nets is not dependent on the dimension of the domain of the regression function unlike kernel methods. Afterwards the comparison moves on more applied aspects and a RBF net joint on an ARCH model is proposed to analyse financial time series.

Translated title of the contribution	[Autom. eng. transl.] RBF neural networks: links and differences with kernel methods.
Original language	Italian
Title of host publication	Atti del convegno MAF 2004 - Metodi matematici e statistici per l'analisi dei dati assicurativi e finanziari
Pages	163-168
Number of pages	6
Publication status	Published - 2004
Event	MAF 2004 - Metodi matematici e statistici per l'analisi dei dati assicurativi e finanziari - Università degli studi di Salerno Duration: 15 Apr 2004 → 16 Apr 2004

Conference

Conference	MAF 2004 - Metodi matematici e statistici per l'analisi dei dati assicurativi e finanziari
City	Università degli studi di Salerno
Period	15/4/04 → 16/4/04

Keywords

ARCH models
Radial Basis Function neural networks
nonparametric regression

Cite this

@inproceedings{2a5d99b1694d41629138d9babf9a71f3,

title = "Le reti neurali RBF: legami e differenze con i metodi kernel.",

abstract = "The Radial Basis Function (RBF) neural networks are nonparametric regression tools similar in formula to the Nadaraya-Watson kernel estimator. Starting from this resemblance, the relationships between the RBF neural nets and the kernel methods are discussed. Some results regarding the moment matrix of the radial basis functions are used to understand the linkage between these estimators and to show a different behaviour of their bandwidth. It is also recalled that the rate of convergence of RBF nets is not dependent on the dimension of the domain of the regression function unlike kernel methods. Afterwards the comparison moves on more applied aspects and a RBF net joint on an ARCH model is proposed to analyse financial time series.",

keywords = "ARCH models, Radial Basis Function neural networks, nonparametric regression, ARCH models, Radial Basis Function neural networks, nonparametric regression",

author = "Mancuso, \{Diego Attilio\}",

year = "2004",

language = "Italian",

isbn = "88-901355-0-6",

pages = "163--168",

booktitle = "Atti del convegno MAF 2004 - Metodi matematici e statistici per l'analisi dei dati assicurativi e finanziari",

note = "MAF 2004 - Metodi matematici e statistici per l'analisi dei dati assicurativi e finanziari ; Conference date: 15-04-2004 Through 16-04-2004",

}

TY - GEN

T1 - Le reti neurali RBF: legami e differenze con i metodi kernel.

AU - Mancuso, Diego Attilio

PY - 2004

Y1 - 2004

N2 - The Radial Basis Function (RBF) neural networks are nonparametric regression tools similar in formula to the Nadaraya-Watson kernel estimator. Starting from this resemblance, the relationships between the RBF neural nets and the kernel methods are discussed. Some results regarding the moment matrix of the radial basis functions are used to understand the linkage between these estimators and to show a different behaviour of their bandwidth. It is also recalled that the rate of convergence of RBF nets is not dependent on the dimension of the domain of the regression function unlike kernel methods. Afterwards the comparison moves on more applied aspects and a RBF net joint on an ARCH model is proposed to analyse financial time series.

AB - The Radial Basis Function (RBF) neural networks are nonparametric regression tools similar in formula to the Nadaraya-Watson kernel estimator. Starting from this resemblance, the relationships between the RBF neural nets and the kernel methods are discussed. Some results regarding the moment matrix of the radial basis functions are used to understand the linkage between these estimators and to show a different behaviour of their bandwidth. It is also recalled that the rate of convergence of RBF nets is not dependent on the dimension of the domain of the regression function unlike kernel methods. Afterwards the comparison moves on more applied aspects and a RBF net joint on an ARCH model is proposed to analyse financial time series.

KW - ARCH models

KW - Radial Basis Function neural networks

KW - nonparametric regression

KW - ARCH models

KW - Radial Basis Function neural networks

KW - nonparametric regression

UR - http://hdl.handle.net/10807/21476

M3 - Contributo a conferenza

SN - 88-901355-0-6

SP - 163

EP - 168

BT - Atti del convegno MAF 2004 - Metodi matematici e statistici per l'analisi dei dati assicurativi e finanziari

T2 - MAF 2004 - Metodi matematici e statistici per l'analisi dei dati assicurativi e finanziari

Y2 - 15 April 2004 through 16 April 2004

ER -

Le reti neurali RBF: legami e differenze con i metodi kernel.

Abstract

Conference

Keywords

Other files and links

Fingerprint

Cite this