𝔖 Bobbio Scriptorium
✦   LIBER   ✦

Growing support vector classifiers with controlled complexity

✍ Scribed by E. Parrado-Hernández; I. Mora-Jiménez; J. Arenas-Garcı́a; A.R. Figueiras-Vidal; A. Navia-Vázquez

Publisher
Elsevier Science
Year
2003
Tongue
English
Weight
170 KB
Volume
36
Category
Article
ISSN
0031-3203

No coin nor oath required. For personal study only.

✦ Synopsis

Semiparametric Support Vector Machines have shown to present advantages with respect to nonparametric approaches, in the sense that generalization capability is further improved and the size of the machines is always under control. We propose here an incremental procedure for Growing Support Vector Classiÿers, which serves to avoid an a priori architecture estimation or the application of a pruning mechanism after SVM training. The proposed growing approach also opens up new possibilities for dealing with multi-kernel machines, automatic selection of hyperparameters, and fast classiÿcation methods. The performance of the proposed algorithm and its extensions is evaluated using several benchmark problems.

📜 SIMILAR VOLUMES

Support vector classifiers via gradient
Support vector classifiers via gradient systems with discontinuous righthand sides
✍ Leonardo V. Ferreira; Eugenius Kaszkurewicz; Amit Bhaya 📂 Article 📅 2006 🏛 Elsevier Science 🌐 English ⚖ 852 KB

Gradient dynamical systems with discontinuous righthand sides are designed using Persidskii-type nonsmooth Lyapunov functions to work as support vector machines (SVMs) for the discrimination of nonseparable classes. The gradient systems are obtained from an exact penalty method applied to the constr

A support vector machine classifier with
A support vector machine classifier with automatic confidence and its application to gender classification
✍ Ji Zheng; Bao-Liang Lu 📂 Article 📅 2011 🏛 Elsevier Science 🌐 English ⚖ 711 KB

In this paper, we propose a support vector machine with automatic confidence (SVMAC) for pattern classification. The main contributions of this work to learning machines are twofold. One is that we develop an algorithm for calculating the label confidence value of each training sample. Thus, the lab