Learning algorithm for nonlinear support vector machines suited for digital VLSI

被引：17

作者：

Anguita, D ^{[1
]}

Boni, A ^{[1
]}

Ridella, S ^{[1
]}

机构：

[1] Univ Genoa, Dept Biophys & Elect Engn, I-16145 Genoa, Italy

来源：

ELECTRONICS LETTERS | 1999年 / 35卷 / 16期

关键词：

D O I：

10.1049/el:19990950

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

A learning algorithm for radial basis function support vector machines (RBF-SVMs) that can be easily implemented in digital VLSI is proposed. It is shown that, as opposed to traditional artificial neural networks, learning in SVMs is very robust with respect to quantisation effects deriving from the finite precision of computations.

引用

页码：1349 / 1350

页数：2

共 6 条

[1] Circuital implementation of support vector machines [J].

Anguita, D ;

Ridella, S ;

Rovetta, S .

ELECTRONICS LETTERS, 1998, 34 (16) :1596-1597

[2]

Friess T., 1998, P 15 TH INT C MACHIN

[3] Support vector machines [J].

Hearst, MA .

IEEE INTELLIGENT SYSTEMS & THEIR APPLICATIONS, 1998, 13 (04) :18-21

[4] Characterization of the Sonar Signals Benchmark [J].

Torres Moreno J.M. ;

Gordon M.B. .

Neural Processing Letters, 1998, 7 (1) :1-4

[5] Convergence analysis of a discrete-time recurrent neural network to perform quadratic real optimization with bound constraints [J].

Perez-Ilzarbe, MJ .

IEEE TRANSACTIONS ON NEURAL NETWORKS, 1998, 9 (06) :1344-1351

[6]

VAPNIK VN, 1999, STAT LEARNING TYHEOR

← 1 →