Research and Application of Adaptative Weighted p-Norm LS-SVM

Pei Zhang Xie; Xing Peng Zhou

doi:10.4028/www.scientific.net/AMM.490-491.1692

Paper Titles

An Autonomous Predicted Star Locations Algorithm for Star Sensor
p.1671

Evaluation on Filter Cleaning Performances Measured by VDI and JIS Testing Rigs for Cleanable Fabric Filter
p.1676

The Automatic Measurement of Complex Surface Based on ZOO-Type CMM
p.1682

A Calibration Method for Binocular Vision System Based on CPSO-BP Netural Network
p.1686

Research and Application of Adaptative Weighted p-Norm LS-SVM
p.1692

Application of Marginal Spectrum in Active Lamb Wave Damage Monitoring of Plate Structure
p.1698

Translate the Stateflow Models into Alloy for Safety Analysis
p.1702

The Settlement Computation of Composite Foundation of Different Typed-Pile Based on the Load Test
p.1706

Application Seismoelectric Method for Inspection Electrically Conducting Media
p.1712

HomeApplied Mechanics and MaterialsApplied Mechanics and Materials Vols. 490-491Research and Application of Adaptative Weighted...

Research and Application of Adaptative Weighted p-Norm LS-SVM

Abstract:

In order to overcome some disadvantages of LS-SVM, such as noise-sensitive and less sparsity, adaptive weighted p-norm LS-SVM is proposed. Experiments shows that different data, using a different regularization(p) can improve the accuracy of the regression, where 1<p<2. p-norm LS-SVM can be convented to the form of compressed sensing, then it can be solved using IRLS. For each sample, because they are not the same, weighted membership degree is introduced to the optimization function so that leading to less error. There are some parameters of weighted p-norm LS-SVM, genetic algorithm is introduced to obtain the optimal parameters. Case study shows that adaptive weighted p-norm LS-SVM is better than other SVM and good result is obtained.

You might also be interested in these eBooks

View Preview

Info:

Periodical:

Applied Mechanics and Materials (Volumes 490-491)

Pages:

1692-1697

DOI:

https://doi.org/10.4028/www.scientific.net/AMM.490-491.1692

Citation:

Cite this paper

Online since:

January 2014

Authors:

Pei Zhang Xie*, Xing Peng Zhou

Keywords:

Adaptive Weighted p-Norm LS-SVM, Compressed Sensing, Genetic Algorithm (GA), IRLS

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

* - Corresponding Author

References

[1] Zhu J, Rosset S, Hastie T, et al. 1-norm support vector machines[J]. Advances in neural information processing systems, 2004, 16(1): 49-56.

Google Scholar

[2] Zou H. An Improved 1-norm SVM for Simultaneous Classification and Variable Selection[J]. Journal of Machine Learning Research-Proceedings Track, 2007, 2: 675-681.

Google Scholar

[3] Mangasarian O L. Exact 1-norm support vector machines via unconstrained convex differentiable minimization[J]. The Journal of Machine Learning Research, 2006, 7: 1517-1530.

Google Scholar

[4] Bennett K P, Bredensteiner E J. Duality and geometry in SVM classifiers[C]/ICML. 2000: 57-64.

Google Scholar

[5] Bennett K P, Bredensteiner E J. Duality and geometry in SVM classifiers[C]/ICML. 2000: 57-64.

Google Scholar

[6] Chartrand R, Yin W. Iteratively reweighted algorithms for compressive sensing[C]/Acoustics, Speech and Signal Processing, 2008. ICASSP 2008. IEEE International Conference on. IEEE, 2008: 3869-3872.

DOI: 10.1109/icassp.2008.4518498

Google Scholar

[7] Blumensath T, Davies M E. Iterative hard thresholding for compressed sensing[J]. Applied and Computational Harmonic Analysis, 2009, 27(3): 265-274.

DOI: 10.1016/j.acha.2009.04.002

Google Scholar

[8] Herman M A, Strohmer T. High-resolution radar via compressed sensing[J]. Signal Processing, IEEE Transactions on, 2009, 57(6): 2275-2284.

DOI: 10.1109/tsp.2009.2014277

Google Scholar

[9] Elad M. Optimized projections for compressed sensing[J]. Signal Processing, IEEE Transactions on, 2007, 55(12): 5695-5702.

DOI: 10.1109/tsp.2007.900760

Google Scholar