Optoelectronic spiking neural network

V. P. Kozemiako, Oleg K. Kolesnytskyj, T. S. Lischenko, Waldemar Wójcik, A. Sulemenov

2013

Closed access Metadata (CSL-JSON)
DOI 10.1117/12.2019340
Data confidence 85 high
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

Proposed by the authors compact optoelectronic implementation of a spiking neural network uses a matrix of semiconductor lasers. Also proposed the implementation of neural element on bispin-device, which is able to manage bya range of lasers in the array. Described the principles of the network in the training and operation mode. To modify the connections weights during the training mode used the optically controlled transparant with memory. Network parameters evaluated achievable with current technological capabilities.