Proximal gradient method for huberized support vector machine
From MaRDI portal
Publication:2337512
Abstract: The Support Vector Machine (SVM) has been used in a wide variety of classification problems. The original SVM uses the hinge loss function, which is non-differentiable and makes the problem difficult to solve in particular for regularized SVMs, such as with -regularization. This paper considers the Huberized SVM (HSVM), which uses a differentiable approximation of the hinge loss function. We first explore the use of the Proximal Gradient (PG) method to solving binary-class HSVM (B-HSVM) and then generalize it to multi-class HSVM (M-HSVM). Under strong convexity assumptions, we show that our algorithm converges linearly. In addition, we give a finite convergence result about the support of the solution, based on which we further accelerate the algorithm by a two-stage method. We present extensive numerical experiments on both synthetic and real datasets which demonstrate the superiority of our methods over some state-of-the-art methods for both binary- and multi-class SVMs.
Recommendations
- SVM-Maj: a majorization approach to linear support vector machines with different hinge errors
- Sparse proximal support vector machine with a specialized interior-point method
- Huberized multiclass support vector machines for microarray classification
- An efficient augmented Lagrangian method for support vector machine
- Robust Truncated Hinge Loss Support Vector Machines
Cites work
- A block coordinate descent method for regularized multiconvex optimization with applications to nonnegative tensor factorization and completion
- A Fast Iterative Shrinkage-Thresholding Algorithm for Linear Inverse Problems
- A globally convergent algorithm for nonconvex optimization based on block coordinate update
- A hierarchical multiple kernel support vector machine for customer churn prediction using longitudinal behavioral data
- A modified finite Newton method for fast solution of large scale linear SVMs
- Clustering-based ensembles for one-class classification
- Comparison of Discrimination Methods for the Classification of Tumors Using Gene Expression Data
- Fixed-Point Continuation for \ell₁-Minimization: Methodology and Convergence
- scientific article; zbMATH DE number 3624650 (Why is no real title available?)
- scientific article; zbMATH DE number 2107836 (Why is no real title available?)
- scientific article; zbMATH DE number 845714 (Why is no real title available?)
- Huberized multiclass support vector machines for microarray classification
- Introductory lectures on convex optimization. A basic course.
- LIBLINEAR: a library for large linear classification
- Multicategory Support Vector Machines
- On gradients of functions definable in o-minimal structures
- On semi- and subanalytic geometry
- OnL1-Norm Multiclass Support Vector Machines
- Regularization and Variable Selection Via the Elastic Net
- Robust twin support vector machine for pattern classification
- Statistical comparisons of classifiers over multiple data sets
- Support-vector networks
- The doubly regularized support vector machine
- The Łojasiewicz Inequality for Nonsmooth Subanalytic Functions with Applications to Subgradient Dynamical Systems
- Using SeDuMi 1.02, A Matlab toolbox for optimization over symmetric cones
- Value-at-risk support vector machine: stability to outliers
- Variable selection for multicategory SVM via adaptive sup-norm regularization
Cited in
(12)- SVM-Maj: a majorization approach to linear support vector machines with different hinge errors
- A globally convergent algorithm for nonconvex optimization based on block coordinate update
- Proximal operator and optimality conditions for ramp loss SVM
- Hybrid Jacobian and Gauss-Seidel proximal block coordinate update methods for linearly constrained convex programming
- Accelerated first-order primal-dual proximal methods for linearly constrained composite convex programming
- scientific article; zbMATH DE number 7338724 (Why is no real title available?)
- Analysis of loss functions in support vector machines
- Gradient projection methods for quadratic programs and applications in training support vector machines
- A unified formulation and fast accelerated proximal gradient method for classification
- Reducing the Complexity of Two Classes of Optimization Problems by Inexact Accelerated Proximal Gradient Method
- A majorization penalty method for SVM with sparse constraint
- Multi-task support vector machine classifier with generalized Huber loss
This page was built for publication: Proximal gradient method for huberized support vector machine
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2337512)
