Learning noisy functions via interval models

被引：4

作者：

Calafiore, Giuseppe Carlo ^{[1
]}

机构：

[1] Politecn Torino, Dipartimento Automat & Informat, I-10129 Turin, Italy

来源：

SYSTEMS & CONTROL LETTERS | 2010年 / 59卷 / 07期

关键词：

Set-valued models; Interval approximation; Convex optimization; Identification; Statistical learning; RANDOMIZED SOLUTIONS; IDENTIFICATION;

D O I：

10.1016/j.sysconle.2010.05.003

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

This paper considers the problem of identification of an interval model for an unknown static function using a finite batch of stochastic input-output data {u((i)), y((i))}, i = 1,..., N. The criterion used for identification is that the width of the interval output of the model should be minimized, while containing a given fraction of the observed outputs y((i)). We show that, for suitable finite N. the resulting model will be reliable, that is it will explain any other unseen output, up to a given and arbitrary high probability. (C) 2010 Elsevier B.V. All rights reserved.

引用

页码：404 / 413

页数：10

共 26 条

[21] Set membership prediction of nonlinear time series
Milanese, M
Novara, C
[J]. IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 2005, 50 (11) : 1655 - 1669
[22] Set membership identification of nonlinear systems
Milanese, M
Novara, C
[J]. AUTOMATICA, 2004, 40 (06) : 957 - 975
[23] Soderstrom T., 1989, SYSTEM IDENTIFICATIO
[24] Vapnik V, 1997, ADV NEUR IN, V9, P281
[25] Vapnik V., 1998, Statistical Learning Theory, P5
[26] WAVELET NEURAL NETWORKS FOR FUNCTION LEARNING
ZHANG, J
WALTER, GG
MIAO, YB
LEE, WNW
[J]. IEEE TRANSACTIONS ON SIGNAL PROCESSING, 1995, 43 (06) : 1485 - 1497

← 1 2 3 →