Exact simplification of support vector solutions

Authors:
Tom Downs;Kevin E. Gates;Annette Masters
Affiliations:
School of Information Technology and Electrical Engineering, University of Queensland, Brisbane, Q. 4072, Australia;School of Information Technology and Electrical Engineering and Department of Mathematics, University of Queensland, Brisbane, Q. 4072, Australia;Department of Mathematics, University of Queensland, Brisbane, Q. 4072, Australia
Venue:
The Journal of Machine Learning Research
Year:
2002

Citing 2
Cited 42

Fast training of support vector machines using sequential minimal optimization

Advances in kernel methods
SVMTorch: support vector machines for large-scale regression problems

The Journal of Machine Learning Research

Sparse Online Greedy Support Vector Regression

ECML '02 Proceedings of the 13th European Conference on Machine Learning
Sparseness of support vector machines

The Journal of Machine Learning Research
Fast exact leave-one-out cross-validation of sparse least-squares support vector machines

Neural Networks
Efficient support vector classifiers for named entity recognition

COLING '02 Proceedings of the 19th international conference on Computational linguistics - Volume 1
The cross entropy method for classification

ICML '05 Proceedings of the 22nd international conference on Machine learning
Adaptive simplification of solution for support vector machine

Pattern Recognition
Building Support Vector Machines with Reduced Classifier Complexity

The Journal of Machine Learning Research
Kernel-based online machine learning and support vector reduction

Neurocomputing
The projectron: a bounded kernel-based Perceptron

Proceedings of the 25th international conference on Machine learning
Increasing classification efficiency with multiple mirror classifiers

Expert Systems with Applications: An International Journal
Recursive reduced least squares support vector regression

Pattern Recognition
Data preparation for sample-based face detection

International Journal of Computer Applications in Technology
Letters: Adaptive binary tree for fast SVM multiclass classification

Neurocomputing
Constructing sparse kernel machines using attractors

IEEE Transactions on Neural Networks
On-line independent support vector machines

Pattern Recognition
Prune support vector machines by an iterative process

International Journal of Computers and Applications
On the sparseness of 1-norm support vector machines

Neural Networks
A hybrid optimization strategy for simplifying the solutions of support vector machines

Pattern Recognition Letters
Bounded Kernel-Based Online Learning

The Journal of Machine Learning Research
Classifier complexity reduction by support vector pruning in kernel matrix learning

IWANN'07 Proceedings of the 9th international work conference on Artificial neural networks
The more you learn, the less you store: Memory-controlled incremental SVM for visual place recognition

Image and Vision Computing
The minimum redundancy-maximum relevance approach to building sparse support vector machines

IDEAL'09 Proceedings of the 10th international conference on Intelligent data engineering and automated learning
Multiclass support vector machines for adaptation in MIMO-OFDM wireless systems

Allerton'09 Proceedings of the 47th annual Allerton conference on Communication, control, and computing
Efficient reduction of support vectors in kernel-based methods

ICIP'09 Proceedings of the 16th IEEE international conference on Image processing
An effective method of pruning support vector machine classifiers

IEEE Transactions on Neural Networks
Sparse learning for support vector classification

Pattern Recognition Letters
A hybrid SVM based decision tree

Pattern Recognition
Example-dependent basis vector selection for kernel-based classifiers

ECML PKDD'10 Proceedings of the 2010 European conference on Machine learning and knowledge discovery in databases: Part III
Fast Multiclass SVM Classification Using Decision Tree Based One-Against-All Method

Neural Processing Letters
A novel heuristic for building reduced-set SVMs using the self-organizing map

IWANN'11 Proceedings of the 11th international conference on Artificial neural networks conference on Advances in computational intelligence - Volume Part I
Simplifying SVM with weighted LVQ algorithm

IDEAL'11 Proceedings of the 12th international conference on Intelligent data engineering and automated learning
A kernel-based Perceptron with dynamic memory

Neural Networks
A speedup method for SVM decision

SSPR'06/SPR'06 Proceedings of the 2006 joint IAPR international conference on Structural, Syntactic, and Statistical Pattern Recognition
Cooperative clustering for training SVMs

ISNN'06 Proceedings of the Third international conference on Advances in Neural Networks - Volume Part I
Simplify decision function of reduced support vector machines

MICAI'05 Proceedings of the 4th Mexican international conference on Advances in Artificial Intelligence
An online framework for learning novel concepts over multiple cues

ACCV'09 Proceedings of the 9th Asian conference on Computer Vision - Volume Part I
The support vector tree

Algorithms and Applications
Hierarchical linear support vector machine

Pattern Recognition
A sequential algorithm for sparse support vector classifiers

Pattern Recognition
Large-scale visual concept detection with explicit kernel maps and power mean SVM

Proceedings of the 3rd ACM conference on International conference on multimedia retrieval
Fast pruning superfluous support vectors in SVMs

Pattern Recognition Letters
Training sparse SVM on the core sets of fitting-planes

Neurocomputing

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper demonstrates that standard algorithms for training support vector machines generally produce solutions with a greater number of support vectors than are strictly necessary. An algorithm is presented that allows unnecessary support vectors to be recognized and eliminated while leaving the solution otherwise unchanged. The algorithm is applied to a variety of benchmark data sets (for both classification and regression) and in most cases the procedure leads to a reduction in the number of support vectors. In some cases the reduction is substantial.