Sparse LS-SVMs with L0 - norm minimization
نویسندگان
چکیده
Least-Squares Support Vector Machines (LS-SVMs) have been successfully applied in many classification and regression tasks. Their main drawback is the lack of sparseness of the final models. Thus, a procedure to sparsify LS-SVMs is a frequent desideratum. In this paper, we adapt to the LS-SVM case a recent work for sparsifying classical SVM classifiers, which is based on an iterative approximation to the L0-norm. Experiments on real-world classification and regression datasets illustrate that this adaptation achieves very sparse models, without significant loss of accuracy compared to standard LS-SVMs or SVMs.
منابع مشابه
Quantitative susceptibility imaging with homotopic L0 minimization programming: preliminary study of brain
INTRODUCTION Susceptibility-weighted imaging (SWI) technique is used for neuroimaging to improve visibility of iron deposits, veins, and hemorrhage [1]. Quantitative susceptibility imaging (QSI) improves upon SWI by measuring iron in tissues, which can be useful for molecular/cellular imaging to analyze brain function, diagnose neurological diseases, and quantify contrast agent concentrations. ...
متن کاملHomotopic l0 minimization technique applied to dynamic cardiac MR imaging
Introduction: The l1 minimization technique has been empirically demonstrated to exactly recover an S-sparse signal with about 3S-5S measurements [1]. In order to get exact reconstruction with smaller number of measurements, recently, for static images, Trzasko [2] has proposed homotopic l0 minimization technique. Instead of minimizing the l0 norm which achieves best possible theoretical bound ...
متن کاملArbitrary Norm Support Vector Machines
Support vector machines (SVM) are state-of-the-art classifiers. Typically L2-norm or L1-norm is adopted as a regularization term in SVMs, while other norm-based SVMs, for example, the L0-norm SVM or even the L(infinity)-norm SVM, are rarely seen in the literature. The major reason is that L0-norm describes a discontinuous and nonconvex term, leading to a combinatorially NP-hard optimization pro...
متن کاملThe Nonnegative Zero-Norm Minimization Under Generalized Z-Matrix Measurement
In this paper, we consider the l0 norm minimization problem with linear equation and nonnegativity constraints. By introducing the concept of generalized Z-matrix for a rectangular matrix, we show that this l0 norm minimization with such a kind of measurement matrices and nonnegative observations can be exactly solved via the corresponding lp (0 < p ≤ 1) norm minimization. Moreover, the lower b...
متن کاملA Hybrid L0-L1 Minimization Algorithm for Compressed Sensing MRI
INTRODUCTION Both L1 minimization [1] and homotopic L0 minimization [2] techniques have shown success in compressed-sensing MRI reconstruction using reduced k-space data. L1 minimization algorithm is known to usually shrink the magnitude of reconstructions especially for larger coefficients [1, 3] and non-convex penalty used in homotopic L0 minimization is advocated to replace L1 penalty [3]. H...
متن کامل