Abstract
A new method of learning fast two-dimensional orthogonal transformations is considered. Tunable orthogonal transformations are regarded as special neural networks. The learning takes a finite number of steps. The learning algorithm does not have the error feedback and is absolutely stable. The method is based on fractal filtering of signals and images. Linguistic models are used to determine the topology and structure of fast transformations. Examples are given.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Andrews, H.C., Caspari, K.L.: A general techniques for spectral analysis. IEEE. Tr. Comput. C19, 16–25 (1970)
Solodovnikov, A.I., Spivakovskii, A.M.: Osnovy Teorii I Metody Spektralnoi Obrabotki Informatsii. Leningrad (1986). (in Russian)
Dorogov, A.Y.: Bystrye Neironnye Seti: Proektirovanie, Nastroika, Prilojenie. http://bookfi.org/book/805420. (in Russian)
Good, I.J.: The interaction algorithm and practical fourier analysis. J. R. Statistic. Soc. Ser. B 20, 361–372 (1958)
Dorogov, A.Y., Shestopalov, M.Y.: Neirosetevoe Modelirovanie Regulyarnyh Fraktalov, in Neirocompyutery: razrabotka I primenenie, no. 6 (2007). (in Russian)
Dorogov, A.Y.: Fractal learning of fast orthogonal neural networks. J. Opt. Mem. Neural Netw. (Information Optics) 21, 105–118 (2012)
Acknowledgements
The paper was prepared in SPbETU and is supported by the Contract № 02.G25.31.0149 dated 01.12.2015 (Board of Education of Russia).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2016 Springer International Publishing Switzerland
About this paper
Cite this paper
Dorogov, A.Y. (2016). Two-Dimensional Fast Orthogonal Neural Networks. In: Cheng, L., Liu, Q., Ronzhin, A. (eds) Advances in Neural Networks – ISNN 2016. ISNN 2016. Lecture Notes in Computer Science(), vol 9719. Springer, Cham. https://doi.org/10.1007/978-3-319-40663-3_24
Download citation
DOI: https://doi.org/10.1007/978-3-319-40663-3_24
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-40662-6
Online ISBN: 978-3-319-40663-3
eBook Packages: Computer ScienceComputer Science (R0)