Abstract
Complex layers of formal neurons can be designed on data sets built from a relatively small number of multidimensional feature vectors. Data sets with this structure can almost always be separated linearly. Genetic data sets typically have this property.
Maximizing the margins is a fundamental principle when designing linear classifiers (formal neurons) based on training sets consisting of a small number of multivariate feature vectors. Maximizing Euclidean (L2) margins is a basic concept in the support vector machines (SVM) method of classifiers designing. An alternative approach to designing formal neurons may be to maximize the margins based on the L1 norm. The margins of the L1 norm enable the design of complex layers of formal neurons.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Duda, O.R., Hart, P.E., Stork, D.G.: Pattern Classification. J. Wiley, New York (2001)
Bishop, C.M.: Pattern Recognition and Machine Learning. Springer, Heidelberg (2006)
Bobrowski, L.: Data Exploration and Linear Separability, pp. 1-172. Lambert (2019).
Boser, B.E., Guyon, I., Vapnik, V.N.: A training algorithm for optimal margin classifiers. In: Proceedings of the Fifth Annual Workshop of Computational Learning Theory, vol. 5, pp. 144–152. ACM, Pittsburgh (1992).
Bobrowski, L.: Complexes of low dimensional linear classifiers with L1 margins. In: Nguyen, N.T., Chittayasothorn, S., Niyato, D., Trawiński, B. (eds.) ACIIDS 2021. LNCS (LNAI), vol. 12672, pp. 29–40. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-73280-6_3
Bobrowski, L.: Design of piecewise linear classifiers from formal neurons by some basis exchange technique. Pattern Recogn. 24(9), 863–870 (1991)
Bobrowski, L.: Data Mining Based on Convex and Piecewise Linear (CPL) Criterion Functions (in Polish). Bialystok University of Technology Press, Bialystok (2005)
Rosenblatt, F.: Principles of Neurodynamics. Spartan Books, Washington (1962)
Bobrowski, L., Łukaszuk, T.: Relaxed linear separability (RLS) approach to feature (Gene) subset selection. In: Xia, X. (ed.)Selected Works in Bioinformatics, INTECH, pp. 103–118 (2011)
Simonnard, M.: Linear Programming. Prentice – Hall, New York (1966)
Kuncheva, L.: Combining Pattern Classifiers: Methods and Algorithms, 2nd edn. J. Wiley, New Jersey (2014)
Alon, U., et al.: Broad patterns of gene expression revealed by clustering analysis of tumor and normal colon tissues probed by oligonucleotide arrays. Proc. Natl. Acad. Sci. USA 96, 6745–6750 (1999)
Petersen K.: Ergodic Theory (Cambridge Studies in Advanced Mathematics). Cambridge University Press, Cambridge (1990)
Collins, S.C., Mikhaleva, A., Vrcelj, K., et al.: Large-scale neuroanatomical study uncovers 198 gene associations in mouse brain morphogenesis. Nat. Commun. 10, 3465 (2019)
Acknowledgments
The presented study was supported by the grant WZ/WI-IIT/3/2020 from the Bialystok University of Technology and funded from the resources for research by the Polish Ministry of Science and Higher Education.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Bobrowski, L., Łukaszuk, T. (2021). Repeatable Functionalities in Complex Layers of Formal Neurons. In: Iliadis, L., Macintyre, J., Jayne, C., Pimenidis, E. (eds) Proceedings of the 22nd Engineering Applications of Neural Networks Conference. EANN 2021. Proceedings of the International Neural Networks Society, vol 3. Springer, Cham. https://doi.org/10.1007/978-3-030-80568-5_36
Download citation
DOI: https://doi.org/10.1007/978-3-030-80568-5_36
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-80567-8
Online ISBN: 978-3-030-80568-5
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)