Neural networks with dynamic structure using a GA-based learning method

Everett Fall, Hsin Han Chiang

研究成果: 書貢獻/報告類型會議論文篇章

3 引文 斯高帕斯(Scopus)

摘要

Artificial neural networks (NNs) are traditionally designed with distinctly defined layers (input layer, hidden layers, output layer) and accordingly network design techniques and training algorithms are based on this concept of strictly defined layers. In this paper, a new approach to designing neural networks is presented. The structure of the proposed NN is not strictly defined (each neuron may receive input from any other neuron). Instead, the initial network structure can be randomly generated, and traditional methods of training, such as back-propagation, are replaced or augmented by a genetic algorithm (GA). The weighting of each neuron input is encoded genetically to serve as the genes for the GA. By means of the training data provided to the supervised network, the contribution of each neuron in creating a desired output serves as a selection function. Each of the neurons is then modified to store and recall past weightings for possible future use. A simple network is trained to recognize vertical and horizontal lines as a proof of concept.

原文英語
主出版物標題ICNSC 2015 - 2015 IEEE 12th International Conference on Networking, Sensing and Control
發行者Institute of Electrical and Electronics Engineers Inc.
頁面7-12
頁數6
ISBN(電子)9781479980697
DOIs
出版狀態已發佈 - 2015 6月 1
對外發佈
事件2015 12th IEEE International Conference on Networking, Sensing and Control, ICNSC 2015 - Taipei, 臺灣
持續時間: 2015 4月 92015 4月 11

出版系列

名字ICNSC 2015 - 2015 IEEE 12th International Conference on Networking, Sensing and Control

其他

其他2015 12th IEEE International Conference on Networking, Sensing and Control, ICNSC 2015
國家/地區臺灣
城市Taipei
期間2015/04/092015/04/11

ASJC Scopus subject areas

  • 電腦網路與通信
  • 儀器
  • 控制與系統工程

指紋

深入研究「Neural networks with dynamic structure using a GA-based learning method」主題。共同形成了獨特的指紋。

引用此