Development of the large-scale Full Connected Artificial Neural Network

2016 Fiscal Year Final Research Report

• Project/Area Number: 26330292
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Soft computing
• Research Institution: Sendai National College of Technology
• Principal Investigator: Hayakawa Yoshihiro 仙台高等専門学校, 情報システム工学科, 教授 (20250847)
• Project Period (FY): 2014-04-01 – 2017-03-31
• Keywords: 全結合型 / 人工神経回路網 / IDL / DS-NET / GPGPU / 大規模 / 高速計算

Outline of Final Research Achievements

This study is that examination inspects various studies necessary for the realization of large-scale full connected artificial neural network by approach to the combinatorial optimization problem using the artificial neural network as for "the technology development of huge full connected artificial neural network".
Hence, we investigated the performance of the network using the inverse function delay-less model.
In case of a dynamic-static network, we performed the various investigations in consideration of the constitution in the modular neural network in the Dynamic network, and we discussed about the hierarchical network in the static network.

Free Research Field

ニューラルネットワーク