Abstract
We present a discrete-time formulation for the autonomous learning conjecture. The main feature of this formulation is the possibility to apply the autonomous learning scheme to systems in which the errors with respect to target functions are not well-defined for all times. This restriction for the evaluation of functionality is a typical feature in systems that need a finite time interval to process a unit piece of information. We illustrate its application on an artificial neural network with feed-forward architecture for classification and a phase oscillator system with synchronization properties. The main characteristics of the discrete-time formulation are shown by constructing these systems with predefined functions.
Similar content being viewed by others
References
P. Kaluza, A.S. Mikhailov, Phys. Rev. E 90, 030901 (2014)
R. Rojas, Neural networks: a systematic introduction (Springer-Verlag, Berlin, New-York, 1996)
P. Kaluza, E. Urdapilleta, Eur. Phys. J. B 87, 236 (2014)
M. Inoue, K. Kaneko, PLoS Comput. Biol. 9, e1003001 (2013)
P. Kaluza, M. Inoue, Eur. Phys. J. B 89, 156 (2016)
P. Kaluza, A.S. Mikhailov. Eur. Phys. J. B 85, 129 (2012)
A. Schuster, Int. J. Comput. Intell. 4, 88 (2008)
S. Bornholdt, K. Sneppen, Proc. R. Soc. Lond. B 267, 1459 (2000)
P. Kaluza, M. Ipsen, M. Vingron, A.S. Mikhailov. Phys. Rev. E 75, 015101 (2007)
T. Yanagita, A.S. Mikhailov, Phys. Rev. E 81, 031901 (2010)
Y. Kuramoto, Chemical Oscillations, Waves and Turbulence (Springer, New York, 1984)
R. Urbanczik, W. Senn, Nat. Neurosci. 12, 250 (2009)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Bilen, A.M., Kaluza, P. Autonomous learning by simple dynamical systems with a discrete-time formulation. Eur. Phys. J. B 90, 94 (2017). https://doi.org/10.1140/epjb/e2017-70714-7
Received:
Revised:
Published:
DOI: https://doi.org/10.1140/epjb/e2017-70714-7