Weighted least squares support vector machines: robustness and sparse approximation

By Johan A. K. Suykens, Jos De Brabanter, Lukas Lukas, Joos Vandewalle. Neurocomputing, 2002

1,289 citations recorded by the source corpus.

Subfields: Computational Mechanics, Computer Vision and Pattern Recognition, Signal Processing

View Paper: https://www.semanticscholar.org/paper/80fb68a6db94266b6b90a9cd0f887074a9d10068

How researchers see this paper

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

In Computational Mechanics

In Computer Vision and Pattern Recognition

In Signal Processing

Citations in the corpus

Forecasting effective drought index using a wavelet extreme learning machine (W-ELM) model (2017, methodology): “To solve for the LSSVR parameters the Lagrangian multipliers method is used as follows (Suykens et al. 2002; Okkan and Serbes 2012): LðW ; j; e; aÞ ¼ LðW ; eÞ XN t¼1 at W T/ xtð Þ þ jþ et yt ; ð9Þ where /(x) is a nonlinear mapping function [related to the kernel function as shown in (12)] and a 2…”
Software defect prediction using ensemble learning on selected features (2015, background): “TheW-SVMs classifier assigns weights that are inversely proportional to the class occurrence in the dataset [57].”
Benchmarking Least Squares Support Vector Machine Classifiers (2004, methodology): “In this section, we report on the application of LS-SVMs on 20 benchmark datasets (Blake & Merz, 1998), of which a brief description will be included in Section 7.1.”
Online Prediction of Time Series Data With Kernels (2009, methodology): “In batch modes, this problem was addressed by using pruning [30], [31] and fixed-size approaches [17], [32], [33].”
Kernel-Based Reinforcement Learning (2006, methodology): “Recently, some methods to deal with such issues are proposed in literature [9], [10], [ 11 ] and [12].”
On Robustness Properties of Convex Risk Minimization Methods for Pattern Recognition (2004, neutral): “In statistical learning theory the principle of regularized empirical risk minimization based on convex loss functions plays an important role, see Vapnik (1998). One strong argument in favor of such methods is that many classifiers based on convex loss functions are universally consistent under wea”
Contextual Hypergraph Modelling for Salient Object Detection (2013, background): “This is a binary classification problem, which is associated with a cost-sensitive classification objective function [24]:”
Contextual Hypergraph Modeling for Salient Object Detection (2013, background): “This is a binary classification problem, which is associated with a cost-sensitive classification objective function [24]:”
Design and evaluation of SVR, MARS and M5Tree models for 1, 2 and 3-day lead time forecasting of river flow data in a semiarid mountainous catchment (2018, background): “…to kernel function describing the inner product in the D-dimension feature space. k x; yð Þ ¼ XD i¼1 /j xð Þ/i yð Þ ð7Þ Equation (7) shows that any symmetric kernel function k satisfying Mercer’s condition corresponds to a dot product in some feature space (Suykens et al. 2002; Tricomi 1985).”
Comparison of SVM and LS-SVM for Regression (2005, methodology): “The data are heteroscedastic so that it forms a challenging test case in some sense [5].”