Application of complex extreme learning machine to multiclass classification problems with high dimensionality: a THz spectra classification problem

Lists

Tools

Yin, X.-X., Hadjiloucas, S. ORCID: https://orcid.org/0000-0003-2380-6114, He, J., Zhang, Y., Wang, Y. and Zhang, Z. (2015) Application of complex extreme learning machine to multiclass classification problems with high dimensionality: a THz spectra classification problem. Digital Signal Processing, 40. pp. 40-52. ISSN 1051-2004

Full text not archived in this repository.

It is advisable to refer to the publisher's version if you intend to cite from this work. See Guidance on citing.

To link to this item DOI: 10.1016/j.dsp.2015.01.007

Abstract/Summary

We extend extreme learning machine (ELM) classifiers to complex Reproducing Kernel Hilbert Spaces (RKHS) where the input/output variables as well as the optimization variables are complex-valued. A new family of classifiers, called complex-valued ELM (CELM) suitable for complex-valued multiple-input–multiple-output processing is introduced. In the proposed method, the associated Lagrangian is computed using induced RKHS kernels, adopting a Wirtinger calculus approach formulated as a constrained optimization problem similarly to the conventional ELM classifier formulation. When training the CELM, the Karush–Khun–Tuker (KKT) theorem is used to solve the dual optimization problem that consists of satisfying simultaneously smallest training error as well as smallest norm of output weights criteria. The proposed formulation also addresses aspects of quaternary classification within a Clifford algebra context. For 2D complex-valued inputs, user-defined complex-coupled hyper-planes divide the classifier input space into four partitions. For 3D complex-valued inputs, the formulation generates three pairs of complex-coupled hyper-planes through orthogonal projections. The six hyper-planes then divide the 3D space into eight partitions. It is shown that the CELM problem formulation is equivalent to solving six real-valued ELM tasks, which are induced by projecting the chosen complex kernel across the different user-defined coordinate planes. A classification example of powdered samples on the basis of their terahertz spectral signatures is used to demonstrate the advantages of the CELM classifiers compared to their SVM counterparts. The proposed classifiers retain the advantages of their ELM counterparts, in that they can perform multiclass classification with lower computational complexity than SVM classifiers. Furthermore, because of their ability to perform classification tasks fast, the proposed formulations are of interest to real-time applications.

Item Type:	Article
Refereed:	Yes
Divisions:	Life Sciences > School of Biological Sciences > Department of Bio-Engineering
ID Code:	39712
Uncontrolled Keywords:	Complex extreme learning machine Reproducing Kernel Hilbert Space Quaternary classification Lagrangian Multiclass classification
Publisher:	Elsevier

Altmetric

Deposit Details

References

[1]J.M. Leiva-Murillo, A. Artes-Rodríguez, Information-theoretic linear feature ex-traction based on kernel density estimators: a review, IEEE Trans. Syst. Man Cybern., Part C Appl. Rev. 42(6) (2012) 1180–1189. [2]G. Huang, H. Zhou, X. Ding, R. Zhang, Extreme learning machine for regres-sion and multiclass classification, IEEE Trans. Syst. Man Cybern., Part B, Cybern. 42(2) (2011) 513–529. [3]G.-B. Huang, D. Wang, Y. Lan, Extreme learning machines: a survey, Int. J. Mach. Learn. Cybern. 2(2) (2011) 107–122. [4]C. Burges, A tutorial on support vector machines for pattern recognition, Data Min. Knowl. Discov. 2 (1998) 121–167. [5]V. Vapnik, The Nature of Statistical Learning Theory, Springer-Verlag, New York, USA, 1995. [6]X. Yin, B. Ng, B. Fischer, B. Ferguson, D. Abbott, Support vector machine ap-plications in terahertz pulsed signals feature sets, IEEE Sens. J. 7(12) (2007) 1597–1608. [7]X. Yin, B. Ng, D. Abbott, Terahertz Imaging for Biomedical Applications: Pattern Recognition and Tomographic Reconstruction, Springer-Verlag, New York, USA, 2012. [8]Y. Iye, Composite fermions and bosons: an invitation to electron masquerade in Quantum Hall, Proc. Natl. Acad. Sci. USA 96 (1999) 8821–8822. [9]K.R. Waters, J. Mobley, J.G. Miller, Causality-imposed (Kramers–Kronig) rela-tionships between attenuation and dispersion, IEEE Trans. Ultrason. Ferroelectr. Freq. Control 52(5) (2005) 822–833. [10]A. El-Gindy, G. Hadad, Nonparametric Bayeserror estimation using unclassified samples, J. AOAC Int. 95(3) (2012) 609–623. [11]G.-B. Huang, Q.-Y. Zhu, C.-K. Siew, Extreme learning machine: theory and ap-plications, Neurocomputing 70(1–3) (2006) 489–501. [12]P. Bouboulis, K. Slavakis, S. Theodoridis, Adaptive learning in complex repro-ducing kernel Hilbertspaces employing Wirtinger’ssubgradients, IEEE Trans. Neural Netw. Learn. Syst. 2(99) (2012) 260–276. [13]G. Scheuermann, H. Krüger, M. Menzel, A. Rockwood, Visualizing nonlinear vec-tor field topology, IEEE Trans. Vis. Comput. Graph. 4(2) (1998) 109–116. [14]H. Li, Invariant Algebras and Geometric Reasoning, World Scientific, Singapore, 2008. [15]T. Nitta, Complex-Valued Neural Networks: Utilizing High-Dimensional Param-eters, Information Science Reference, Pennsylvania, 2009. [16]E. Hitzer, T. Nitta, Y. Kuroe, Applications of Clifford’sgeometric algebra, Adv. Appl. Clifford Algebras 23(2) (2013) 377–404. [17]V. Dietrich, K. Habetha, G. Jank, Clifford Algebras and Their Application in Mathematical Physics, Kluwer Academic Publishers, Springer, Dordrecht, The Netherlands, 1998. [18]E. Corrochano, G. Sobczyk, Geometric Algebra with Applications in Science and Engineering, Birkhäuser Boston, c/o Springer-Verlag, New York, 2001. [19]B. Sethuraman, Division algebras and wireless communication, Not. Am. Math. Soc. 57(11) (2010) 1432–1439. [20]A. Hirose, Complex-Valued Neural Networks, World Scientific Publishing, Sin-gapore, 2003. [21]E. Bayro-Corrochano, Geometric neural computing, IEEE Trans. Neural Netw. 12(5) (2001) 968–986. [22]E. Bayro-Corrochano, N. Arana-Daniel, Clifford support vector machines for classification, regression, and recurrence, IEEE Trans. Neural Netw. 21(11) (2010) 1731–1746. [23]M.-B. Li, G.-B. Huang, P. Saratchandran, N. Sundararajan, Fully complex extreme learning machine, Neurocomputing 68 (2005) 306–314. [24]R.V. Babu, S. Suresh, Fully complex-valued ELM classifiers for human ac-tion, in: The 2011 International Joint Conference on Neural Networks, 2011, pp.2803–2808. [25]G.C.W.J.W. Bowen, S. Hadjiloucas, Sample-induced beam distortions in tera-hertz time domain spectroscopy and imaging systems, in: Joint 32nd Inter-national Conference on Infrared and Millimetre Waves and 15th International Conference on Terahertz Electronics, 2007, pp.208–209. [26]S. Hadjiloucas, L. Karatzas, J. Bowen, Measurements of leaf water content using terahertz radiation, IEEE Trans. Microw. Theory Tech. 47(2) (1999) 142–149. [27]F.A. Tobar, A. Kuh, D.P. Mandic, A novel augmented complex valued kernel LMS, in: The7th IEEE Sensor Array and Multichannel Workshop, 2012, pp.473–476. [28]A. Shilton, D. Lai, Quaternionic and complex-valued support vector regression for equalization and function approximation, in: International Joint Conference on Neural Networks, IJCNN, 2007, pp.920–925. [29]P. Bouboulis, S. Theodoridis, Extension of Wirtinger’s calculus to reproducing kernel Hilbert spaces and the complex kernel LMS, IEEE Trans. Signal Process. 59(3) (2011) 964–978. [30]G.-B. Huang, H. Zhou, X. Ding, R. Zhang, Extreme learning machine for regres-sion and multiclass classification, IEEE Trans. Syst. Man Cybern., Part B, Cybern. 42(2) (2012) 513–529. [31]X.-X. Yin, B. Ng, B. Ferguson, D. Abbott, Wavelet based local tomographic image using terahertz techniques, Digit. Signal Process. 19(4) (2009) 750–763. [32]B. Ferguson, S. Wang, H. Zhong, D. Abbott, X.-C. Zhang, Powder retection with t-ray imaging, in: Terahertz for Military and Security Application, in: Proceed-ing of SPIE, vol.5070, 2003, pp.7–16. [33]K. Fukunaga, D.L. Kessell, Chemometrics in pharmaceutical analysis: an in-troduction, review, and future perspectives, IEEE Trans. Inf. Theory IT-19(4) (1973) 434–440. [34]K. Fukunaga, D.M. Hummels, Leave-one-out procedures for nonparametric error estimates, IEEE Trans. Pattern Anal. Mach. Intell. 11(4) (1989) 421–423. [35]R.R. Jackson, C.M. Carter, M.S. Tarsitano, Trial-and-error solving of a confine-ment problem by a jumping spider, Portiafimbriata, Behaviour 138(10) (2001) 1215–1234. [36]Y.J.Y. Jin, B. Sendhoff, Pareto-based multi-objective machine learning: an overview and case studies, IEEE Trans. Syst. Man Cybern., Part C Appl. Rev. 38(3) (2008) 397–415. [37]J.D. Rodríguez, A. Perez, D. Arteta, D. Tejedor, J.A. Lozano, Using multidimen-sional Bayesiannetwork classifiers to assist the treatment of multiple sclerosis, IEEE Trans. Syst. Man Cybern., Part C Appl. Rev. 42(6) (2012) 1705–1715. [38]H.Q. Vu, G. Li, N.S. Sukhorukova, G. Beliakov, S. Liu, C. Philippe, H. Amiel, A.Ugon, K-complex detection using a hybrid-synergic machine learning method, IEEE Trans. Syst. Man Cybern., Part C Appl. Rev. 39(4) (2012) 1478–1490. [39]D. Muni, N. Pal, J. Das, Genetic programming for simultaneous feature selection and classifierdesign, IEEE Trans. Syst. Man Cybern., PartC Appl. Rev. 36(1) (2006) 106–117. [40]K. Mao, Identifying critical variables of principal components for unsupervised feature selection, IEEE Trans. Syst. Man Cybern., Part C Appl. Rev. 35(2) (2005) 339–344. [41]S. Salcedo-Sanz, G. Camps-Valls, F. Perez-Cruz, J. Sepulveda-Sanchis, C. Bousono-Calzon, Enhancing genetic feature selection through restricted search and Walshanalysis, IEEE Trans. Syst. Man Cybern., PartC Appl. Rev. 34(4) (2004) 398–406. [42]J. Park, S. Baek, S. Bae, M. Jeong, Dual features functional support vector ma-chines for fault detection of rechargeable batteries, IEEE Trans. Syst. Man Cy-bern., Part C Appl. Rev. 39(4) (2009) 480–485. [43]O. Omitaomu, M. Jeong, A. Badiru, J. Hines, On-line support vector regression approach for the monitoring of motor shaft misalignment and feed-water low rate, IEEE Trans. Syst. Man Cybern., Part C Appl. Rev. 37(5) (2007) 962–970. [44]P. Maji, S. Paul, Rough sets for selection of molecular descriptors to predict biological activity of molecules, IEEE Trans. Syst. Man Cybern., Part C Appl. Rev. 40(6) (2010) 639–648. [45]G. Duan, Y.-W. Chen, A machine learning-based framework for automatic visual inspection of microdrill bits in PCB production, IEEE Trans. Syst. Man Cybern., Part C Appl. Rev. 42(6) (2012) 1679–1689. [46]Y.-S. Jeong, I.-H. Kang, M.-K. Jeong, D. Kong, A new feature selection method for one-class classification problems, IEEE Trans. Syst. Man Cybern., Part C Appl. Rev. 39(4) (2012) 1500–1509.

University Staff: Request a correction | Centaur Editors: Update this record

University of Reading

CentAUR: Central Archive at the University of Reading

Accessibility navigation

Application of complex extreme learning machine to multiclass classification problems with high dimensionality: a THz spectra classification problem

Abstract/Summary

Page navigation

See also

Footer navigation