Efficient tuning of SVM hyperparameters using radius/margin bound and iterative algorithms

By S. Sathiya Keerthi. IEEE Trans. Neural Networks, 2002

340 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/349d12a77544591c3f4654f51a768e16a84cbbc6

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Efficient multiple hyperparameter learning for log-linear models (2007, methodology): “In SVMs, approaches based on optimizing generalization bounds [4], such as the r adius/margin-bound [15] or maximal discrepancy criterion [2] have been proposed.”
A multiple kernel support vector machine scheme for feature selection and rule extraction from gene expression data of cancer tissue (2007, methodology): “The optimization of the feature coefficients bd can be seen as model selection problem in SVM which has been well studied [51—54].”
Supervised classification and mathematical optimization (2013, methodology): “In [130], the focus is on the radius-margin bound.”
Feature selection via sensitivity analysis of SVM probabilistic outputs (2008, methodology): “Typically, these parameters are obtained using the crossvalidation procedure although other methods have also been discussed (Lee and Lin 2000; Chapelle et al. 2002; Keerthi 2002).”
Evolutionary tuning of multiple SVM parameters (2005, methodology): “These algorithms iterate the following procedure: The SVM is trained using the current hyperparameter vector, the gradient of some generalization error bound w.r.t. the hyperparameters is calculated, and a step is performed in the parameter space based on this gradient.”
Support vector regression with genetic algorithms in forecasting tourism demand (2007, methodology): “To construct the SVR model efficiently, SVR’s parameters must be set carefully (Duan, Keerthi, & Poo, 2001; Keerthi, 2002; Lin, 2001).”
A Pattern Search Method for Model Selection of Support Vector Regression (2002, background): “An efficient implementation for the radius-margin bound for large problems is reported in [8].”
An Experimental Study of the Combination of Meta-Learning with Particle Swarm Algorithms for SVM Parameter Selection (2012, methodology): “In order to evaluate our study, we adapted the prototypes to select two SVM parameters: the parameter γ of the RBF kernel and the regularization constant C, which may have a strong influence in SVM performance [10].”