After having introduced the different possiblities to realize the most important components of optical neural networks — interconnection and storage systems, threshold devices and optical processing configurations — it is time to put these components together to complete optical neural networks. The following chapters will present examples of realizations of the different neural network types following the schematic presented in chapter 2. The emphasize of the next chapters is again on those systems that use the components and algorithms I have described in the first two parts of the book. I have focussed on those realizations that are based on all-optical components, using optoelectronic devices to support the optical function of the system. Neural network implementations using mainly electronic elements that are supported by optical interconnections are only mentioned rarely for some special realizations, but are mainly left out in this part.
KeywordsAssociative Memory Ring Resonator Reference Beam Phase Conjugation Bidirectional Associative Memory
Unable to display preview. Download preview PDF.
- 2.G. Hinton, J. Anderson, Parallel Models of Associative Memories, LawrenceErlbaum, 1981.Google Scholar
- 3.J.J. Hopfield, Neural networks and physical systems with emergent collective computational abilities, Proc. Nat. Acad. of Sciences, 79 (1982)..Google Scholar
- 4.Y. Abu-Mostafa, D. Psaltis, Optical pattern recognition, Scientific American, March 1987.Google Scholar
- 5.K. Kyuma, Optical neural networks a review, Nonlinear Optics 1 (1991), pp. 39–49. A short and compact overview over current implementations of photorefractive neural networks, especially of associative memories.Google Scholar