Methods of parallel vertical data processing in neural networks

1Grytsyk, VV
1Tsmots, IG
1Skorokhoda, OV
1Lviv Polytechnic National University
Dopov. Nac. akad. nauk Ukr. 2014, 10:40-44
https://doi.org/10.15407/dopovidi2014.10.040
Section: Information Science and Cybernetics
Language: Ukrainian
Abstract: 

An operational basis of neural networks has been identified. The feasibility of development of parallel-vertical hardware neural networks has been substantiated. A parallel-vertical data processing method in neural elements (neural networks) that is oriented to the VLSI implementation and provides a reduction of the number of interface's pins, the bitness of interneuron connection, and equipment costs has been developed. The principles of the VLSI implementation of neural networks have been proposed.

Keywords: neural networks, processing methods
References: 

1. Haikin S. Neural networks: a complete course. – Moscow: Williams, 2006. – 1104 p.
2. Tsmots I. G., Skorokhoda O. V., Krasovskyi V. B. Zb. nauk. prats In-ty problem modeliuvannia v energetytsi, 2013, Iss. 66: 149–155 (in Ukrainian).
3. Tkachenko R. O., Tsmots I. G., Skorokhoda O. V. Tekhn. visti, 2010 1(31), 2(32): P. 166–169.
4. Tsmots I. G., Skorokhoda O. V., Baluch B. I. Modeliuvannia ta informatsiini tekhnologii, 2013, Iss. 67: 160–166 (in Ukrainian).
5. Grytsyk V. V., Tsmots I. G., Teslyuk V. M. Dopov. Nac. akad. nauk Ukr., 2013, No 1: 30–36 (in Ukrainian).