Abstract
This paper presents the development of the algorithm for adaptive construction of hierarchical neural network classifiers based on automatic modification of the desired response of a perceptron with a small number of neurons in a single hidden layer. Improved versions of the algorithm are tested on standard benchmark problems Vowels and MNIST. A discussion of the results, strengths and weaknesses of the algorithm, directions of further work on its testing and improvement, is provided.
Similar content being viewed by others
References
Dolenko, S.A., Orlov, Yu.V., Persiantsev, I.G., Shugai J.S., and Eremin, E.K., Adaptive method of construction of a perceptron-based hierarchical structure of classifiers, The 4th Open Russian-German Workshop on Pattern Recognition and Image Analysis (Valdai, Russia, 1996), pp. 45–49.
Dolenko, S.A., Orlov, Yu.V., Persiantsev, I.G., Shugai, Ju.S., and Eremin, E.K., The perceptron-based hierarchical structure of classifiers constructed by the adaptive method, Pattern Recognit. Image Anal., 1997, vol. 7, no. 1, pp. 24–28.
Dolenko, S.A., Orlov, Yu.V., Persiantsev, I.G., Shugai, Ju.S., and Eremin, E.K., Self-organizing construction of hierarchical structure of multi-layer perceptrons, Proceedings Fifth Int. Conf. on Artificial Neural Networks (UK, University of Cambridge, Churchill College, 1997), IEE Conference Publication, 1997, vol. 440, pp. 285–290.
Dolenko, S.A., Orlov, Yu.V., Persiantsev, I.G., and Shugai, Yu.S., Use of self-organization principles in construction of hierarchical neural network classifiers, Pattern Recognit. Image Anal., 1998, vol. 8, no. 2, pp. 122–124.
Orlov, Yu.V., Development and study of the algorithms of construction of neural network complexes for classification of data of physical measurements, Candidate Thesis in Phys. and Math., SINP MSU, 2003 (in Russian).
Persiantsev, I.G., Adaptive construction of hierarchical neural network systems for classification and for segmentation of time series, XII All-Russian Scientific and Technical Conference “Neuroinformatics-2010”: Lectures in Neuroinformatics, Moscow, NIJaU MIFI, 2010, pp. 212–242 (in Russian).
Svetlov, V.A., Persiantsev, I.G., Shugay, J.S., and Dolenko, S.A., A New implementation of the algorithm of adaptive construction of hierarchical neural network classifiers, Opt. Mem. Neural Networks (Inform. Opt.), 2015, vol. 24, no. 4, pp. 288–294.
Utgoff, P.E., Perceptron trees: A case study in hybrid concept representations, Connection Sci., 1989, vol. 1, no. 4, pp. 377–391.
Utgoff, P.E. and Carla E., An incremental method for finding multivariate splits for decision trees, Proceedings of the Seventh International Conference on Machine Learning, 1990, pp. 58–65.
Guo, Heng and Gelfand, S.B., Classification trees with neural network feature extraction, IEEE Trans. Neural Networks, 1992, vol. 3, no. 6, pp. 923–933.
Sakar, A. and Mammone, R.J., Growing and pruning neural tree networks, IEEE Trans. Computers, 1993, vol. 42, no. 3, pp. 291–299.
Song, Hee-Heon and Lee, Seong-Whan, A self-organizing neural tree for large-set pattern classification, IEEE Trans. Neural Networks, 1998, vol. 9, no. 3, pp. 369–380.
Huang, De-Shuang, Guo, Lin, et al., A novel clustering-neural tree for pattern classification, 2004 IEEE International Joint Conference on Neural Networks, Proceedings, 2004, vol. 2, pp. 1303–1308.
Foresti, Gian Luca and Dolso, T., An adaptive high-order neural tree for pattern recognition, IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics, 2004, vol. 34, no. 2, pp. 988–996.
Foresti, Gian Luca and Micheloni, Christian, Generalized neural trees for pattern classification, IEEE Trans. Neural Networks, 2002, vol. 13, no. 6, pp. 1540–1547.
Micheloni, Christian, Rani, Asha, Kumar, Sanjeev, and Foresti, Gian Luca, A balanced neural tree for pattern classification, Neural Networks, 2012, vol. 27, pp. 81–90.
Lichman, M., UCI Machine Learning Repository. Irvine, CA: University of California, School of Information and Computer Science, 2013. http://archive.ics.uci.edu/ml. Connectionist Bench (Vowel Recognition–Deterding Data) Data Set.https://archive.ics.uci.edu/ml/datasets/Connectionist+Bench+(Vowel+Recognition+-+ Deterding+Data). Cited November 21, 2016.
LeCun, Y., Cortes, C., and Burges, C.J.C., The MNIST database of handwritten digits. http://yann.lecun.com/exdb/mnist/. Cited November 21, 2016.
Sutskever, I. et al., On the importance of initialization and momentum in deep learning, ICML, 2013, vol. 28, no. 3, pp. 1139–1147.
Softmax Classification. http://knet.readthedocs.io/en/latest/softmax.html. Cited November 21, 2016.
Author information
Authors and Affiliations
Corresponding author
About this article
Cite this article
Svetlov, V.A., Dolenko, S.A. Development of the algorithm of adaptive construction of hierarchical neural network classifiers. Opt. Mem. Neural Networks 26, 40–46 (2017). https://doi.org/10.3103/S1060992X17010076
Received:
Published:
Issue Date:
DOI: https://doi.org/10.3103/S1060992X17010076