Abstract
This paper presents simple feature quantities for learning of dynamic binary neural networks. The teacher signal is a binary periodic orbit corresponding to control signal of switching circuits. The feature quantities characterize generation of spurious memories and stability of the teacher signal. We present a simple greedy search based algorithm where the two feature quantities are used as cost functions. Performing basic numerical experiments, the algorithm efficiency is confirmed.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Kouzuki, R., Saito, T.: Learning of simple dynamic binary neural networks. IEICE Trans. Fundam. E96–A(8), 1775–1782 (2013)
Moriyasu, J., Saito, T.: Sparsification and stability of simple dynamic binary neural networks. IEICE Trans. Fundam. E97–A(4), 985–988 (2014)
Nakayama, Y., Kouzuki, R., Saito, T.: Application of the dynamic binary neural network to switching circuits. In: Lee, M., Hirose, A., Hou, Z.-G., Kil, R.M. (eds.) ICONIP 2013, Part II. LNCS, vol. 8227, pp. 697–704. Springer, Heidelberg (2013)
Moriyasu, J., Saito, T.: A deep dynamic binary neural network and its application to matrix converters. In: Wermter, S., Weber, C., Duch, W., Honkela, T., Koprinkova-Hristova, P., Magg, S., Palm, G., Villa, A.E.P. (eds.) ICANN 2014. LNCS, vol. 8681, pp. 611–618. Springer, Heidelberg (2014)
Moriyasu, J., Saito, T.: A cascade system of simple dynamic binary neural networks and its sparsification. In: Loo, C.K., Yap, K.S., Wong, K.W., Teoh, A., Huang, K. (eds.) ICONIP 2014, Part I. LNCS, vol. 8834, pp. 231–238. Springer, Heidelberg (2014)
Gray, D.L., Michel, A.N.: A training algorithm for binary feed forward neural networks. IEEE Trans. Neural Netw. 3(2), 176–194 (1992)
Chen, F., Chen, G., He, Q., He, G., Xu, X.: Universal perceptron and DNA-like learning algorithm for binary neural networks: non-LSBF implementation. IEEE Trans. Neural Netw. 20(8), 1293–1301 (2009)
Chua, L.O.: A Nonlinear Dynamics Perspective of Wolfram’s New Kind of Science, I and II. World Scientific, Singapore (2005)
Wada, W., Kuroiwa, J., Nara, S.: Completely reproducible description of digital sound data with cellular automata. Phys. Lett. A 306, 110–115 (2002)
Rosin, P.L.: Training cellular automata for image processing. IEEE Trans. Image Process. 15(7), 2076–2087 (2006)
Iguchi, T., Hirata, A., Torikai, H.: Theoretical and heuristic synthesis of digital spiking neurons for spike-pattern-division multiplexing. IEICE Trans. Fundam. E93–A(8), 1486–1496 (2010)
Bose, B.K.: Neural network applications in power electronics and motor drives - an introduction and perspective. IEEE Trans. Ind. Electron. 54(1), 14–33 (2007)
Rodriguez, J., Rivera, M., Kolar, J.W., Wheeler, P.W.: A review of control and modulation methods for matrix converters. ieee tie 59(1), 58–70 (2012)
Araki, K., Saito, T.: An associative memory including time-variant self-feedback. Neural Netw. 7(8), 1267–1271 (1994)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2015 Springer International Publishing Switzerland
About this paper
Cite this paper
Sato, R., Saito, T. (2015). Simple Feature Quantities for Learning of Dynamic Binary Neural Networks. In: Arik, S., Huang, T., Lai, W., Liu, Q. (eds) Neural Information Processing. ICONIP 2015. Lecture Notes in Computer Science(), vol 9489. Springer, Cham. https://doi.org/10.1007/978-3-319-26532-2_25
Download citation
DOI: https://doi.org/10.1007/978-3-319-26532-2_25
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-26531-5
Online ISBN: 978-3-319-26532-2
eBook Packages: Computer ScienceComputer Science (R0)