Dimensionality Reduction via Sparse Support Vector Machines

By Jinbo Bi, Kristin P. Bennett, Mark J. Embrechts, Curt M. Breneman, Minghu Song. 2003

492 citations recorded by the source corpus.

How researchers see this paper

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

Citations in the corpus

Random Feature Maps for Dot Product Kernels (2012, background): “There has been a lot of research on SVM formulations with sparsity pro-a moting regularizers (see for example Bi et al., 2003) and support vector reduction (see for example Cos-salter et al., 2011).”
An introduction to variable and feature selection (2003, methodology): “Another paper in the issue, by Bi et al. (2003), uses 1-norm SVMs, without iterative multiplicative updates.”
Machine Learning Methods for Attack Detection in the Smart Grid (2015, methodology): “SVM [48] and kernel machines [49] can be employed.”
A Feature Selection Newton Method for Support Vector Machine Classification (2004, background): “An excellent overview of the state-of-the-art methods for feature selection via support vector machines appears in [2].”
Correlation and variable importance in random forests (2013, methodology): “2004; Díaz-Uriarte and Alvarez de Andrés 2006; Louw and Steel 2006; Genuer et al. (2010)). A classical issue of variable selection methods is their instability: a small perturbation of the training sample may completely change the set of selected variables. This instability is a consequence of the d”
Ultrahigh dimensional feature selection: beyond the linear model (2009, background): “See also the special issue published by JMLR on “variable and feature selection”, including Bi et al. (2003), Bengio & Chapados (2003) and Guyon & Elisseeff (2003).”
Feature selection based on artificial bee colony and gradient boosting decision tree (2019, background): “Bi et al. [24] use l1-norm SVMs, without iterative multiplicative updates, which takes in the context of least-square regression and eliminates features by setting their weights to zero.”
Non-parametric prediction and mapping of standing timber volume and biomass in a temperate forest: application of multiple optical/LiDAR-derived predictors (2010, methodology): “number of chromosomes in replicated runs) of 3000 was set (Bi et al., 2003).”
Support Vector Regression (2007, methodology): “A methodology has been described for performing variable ranking and selection using support vector machines [69].”
Relevant Feature Set Estimation with a Knock-out Strategy and Random Forests (2015, unknown): “…1996) 3 Elastic Net (Zou and Hastie, 2005) 4 Sparse logistic regression (Yamashita et al., 2008) (Sparse LR) 5 Sparse support vector classification (Bi et al., 2003) (Sparse SVC) 6 Random Forest with Gini Contrast (Langs et al., 2011) (RF Gini) 7 Stability Selection based on LASSO (Meinshausen and…”
Sparse logistic regression for whole brain classification of fMRI data (2010, methodology): “SVM based feature selection using L1, L2 or L0 regularization methods was also proposed in the literature (Bi et al., 2003; Perkins et al., 2003; Weston et al., 2003).”
An Introduction to Variable and Feature Selection (2003, unknown): “Another paper in the issue, by Bi et al. (2003), uses ` 1 -norm SVMs, without iterative multiplicative updates.”
Unsupervised feature selection for multi-cluster data (2010, background): “Several supervised feature selection algorithms [2] have been designed to address this issue.”
A Statistical Approach to Material Classification Using Image Patch Exemplars (2009, unknown): “signed perpendicular distance [4] (analogous results hold”
Genetic programming for simultaneous feature selection and classifier design (2006, methodology): “Authors in [24] discuss FS and feature ranking using SVMs.”
MILES: Multiple-Instance Learning via Embedded Instance Selection (2006, methodology): “Thus, 1-norm SVM is also referred to as sparse SVM and has been applied to other practical problems such as drug discovery [8].”
Supersparse Linear Integer Models for Interpretable Classification (2013, methodology): “Other methods to produce sparse linear models with real coeﬃcients include those of Tipping [67], Bi et al. [6], Neylon [54], Giacobello et al. [26], Bazerque and Giannakis [48], and Balakrishnan and Madigan [5].”