HARDWARE IMPLEMENTATION OF DYNAMICAL NEURAL NETWORKS SUITABLE FOR ONLINE TRAINING

V. V. Khilkevichv, Moscow Power Engineering Institute, Moscow, Russia

SYNCHROINFO JOURNAL. Volume 5, Number 3 (2019). P. 22-25.

Abstract

The report is devoted to the hardware implementation of dynamical neural network using Field Programmable Gate Arrays. Simulation results are presented.

Keywords: dynamical neural networks, neural network training, Field Programmable Gate Arrays.

References

[1] R. Beer. On the dynamics of small continuous time recurrent neural networks, Adaptive Behavior, vol. 3, no. 4, pp. 469-509, 1995.
[2] K. Funahashi and Y. Nakamura, Approximation of dynamical systems by continuous time recurrent neural networks, Neural Networks, vol. 6, pp. 801-806, 1993.
[3] R. Beer and J. Gallagher, Evolving dynamical neural networks for adaptive behavior,” Adaptive Behavior, vol. 1, no. 1, pp. 91-122, 1992.
[4] J. Gallagher and J. Fiore, Continuous time recurrent neural networks: A paradigm for evolvable analog controller circuits,” in The Proceedings of the National Aerospace and Electronics Conference. NAECON 2000, pp. 299-304, IEEE Press, 2000.
[5] A. Stoica, C. Lazaro, D. Keymeulen, and K. Hayworth, Evolution of CMOS circuits in simulations and directly in hardware on a programmable chip, in GECCO-99: Proceedings of the Genetic and Evolutionary Computation Conference, vol. 2, pp. 1198-1203, 13-17 July 1999.