FUNDAMENTAL DEVICE AND CIRCUITS FOR SYNAPTIC CONNECTIONS IN SELF-ORGANIZING NEURAL NETWORKS

Using fundamental device and circuits, we have realized three function s required for synaptic connections in self-organizing neural networks : long term memory of synaptic weights, fixed total amount of synaptic weights in a neuron, and lateral inhibition. The first two functions have been condensed into an optical adaptive device and circuits with Boating gates. Lateral inhibition has been realized by a winner-take-a ll circuit and a following lateral excitatory connection circuit. We h ave fabricated these devices and circuits using CMOS technology and co nfirmed the three functions. In addition, topological mapping, which i s essential for feature extraction, has been formed in a primitive net work constructed with the fundamental device and circuits.