Advanced Search

Query:

Fill in the form and click »Search«...

Format:

Display: entries per page entries

Zbl 1075.68632
Cherkassky, Vladimir; Ma, Yunqian
Practical selection of SVM parameters and noise estimation for SVM regression. (English)
[J] Neural Netw. 17, No. 1, 113-126 (2004). ISSN 0893-6080

Summary: We investigate practical selection of hyper-parameters for support vector machines (SVM) regression (that is, $\epsilon$-insensitive zone and regularization parameter $C$). The proposed methodology advocates analytic parameter selection directly from the training data, rather than re-sampling approaches commonly used in SVM applications. In particular, we describe a new analytical prescription for setting the value of insensitive zone $\epsilon$, as a function of training sample size. Good generalization performance of the proposed parameter selection is demonstrated empirically using several low- and high-dimensional regression problems. Further, we point out the importance of Vapnik's $\epsilon$-insensitive loss for regression problems with finite samples. To this end, we compare generalization performance of SVM regression (using proposed selection of $\epsilon$-values) with regression using `least-modulus' loss ($\epsilon=0$) and standard squared loss. These comparisons indicate superior generalization performance of SVM regression under sparse sample settings, for various types of additive noise.

MSC 2000:: *68T05 Learning and adaptive systems

Keywords: Complexity control; Loss function; Parameter selection; Prediction accuracy; Support vector machine regression; VC theory

Comment on this Item PDF MathML XML ASCII DVI PS BibTeX Online Ordering Article Journal

	Scientific prize winners of the ICM 2010
	Overhang
	Lie groups, physics and geometry. An introduction for physicists, engineers and chemists.

Advanced Search

Zentralblatt MATH Berlin [Germany]