Abstract
Most, if not all, optical hardware-based neural networks are slow during the neural learning phase. This limitation has been not only a speed bottleneck, but it has contributed to the lack of wide-spread use of optical neural systems. We present a novel solution – Optical Fixed-Weight Learning Neural Networks. Standard neural networks learn new function mappings by the changing of their synaptic weights. However, the Fixed-Weight Neural Networks learn new mappings by dynamically changing recurrent neural signals. The (fixed) synaptic weights of the FWL-NN implement a learning ”algorithm” which adjusts the recurrent signals toward their proper values.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Keller, P.E., Gmitro, A.F.: Operational Parameters of an Opto-Electronic Neural Network Employing Fixed-Planar Holographic Interconnects. World Congress on Neural Networks (1993)
Abu-Mostafa, Y.S., Psaltis, D.: Optical Neural Computers Scientific American (March 1987)
Kubler, C., Ehrke, H., Huber, R., Lopez, R., Halabica, A., Haglund, R.F., Leiterstorfer, A.: Coherent Structural Dynamics and Electronic Correlations during an Ultrafast Insulator-to-Matal Phase Transition in VO2. Physical Review Letters. PRL 99, 116401 14 (September 2007)
Cotter, N.E., Conwell, P.R.: Learning algorithms and fixed dynamics. In: Proceedings of the International Conference on Neural Networks 1991, vol. I, pp. 799–804. IEEE, Los Alamitos (1991)
Feldkamp, L.A., Prokhorov, D.V., Feldkamp, T.: Conditioned Adaptive Behavior from Kalman Filter Trained Recurrent Networks. IEEE, Los Alamitos (2003)
Younger, A.S., Conwell, P.R., Cotter, N.E.: Fixed-Weight On-Line Learning. IEEE Transactions on Neural Networks 10(2), 272–283 (1999)
Prokhorov, D.V., Feldkamp, L.A., Tyukin, I.Y.: Adaptive Behavior with Fixed Weights in RNN: An Overview. In: IJCNN 2002. IEEE, Los Alamitos (2002)
Lo, J.T., Bassu, D.: Adaptive vs. Accomodative Neural Networks for Adaptive System Identification. In: IJCNN 2001, IEEE, Los Alamitos (2001)
Hochreiter, S., Younger, A.S., Conwell, P.R.: Learning To Learn Using Gradient Descent. In: Proceedings of the International Conference on Artificial Neural Networks, Springer, Heidelberg (2001)
Bade, S.L., Hutchings, B.L.: FPGA-Based Stochastic Neural Networks. In: Implementation IEEE FPGAs for Custom Computing Machines Workshop, Napa, CA, pp. 189–198 (1994)
Werbos, P.: Private Communication (2004)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2008 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Younger, A.S., Redd, E. (2008). Learning at the Speed of Light: A New Type of Optical Neural Network. In: Dolev, S., Haist, T., Oltean, M. (eds) Optical SuperComputing. OSC 2008. Lecture Notes in Computer Science, vol 5172. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-85673-3_9
Download citation
DOI: https://doi.org/10.1007/978-3-540-85673-3_9
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-85672-6
Online ISBN: 978-3-540-85673-3
eBook Packages: Computer ScienceComputer Science (R0)