期刊名称：Brain. Broad Research in Artificial Intelligence and Neuroscience
印刷版ISSN：2067-3957
出版年度：2010
卷号：1
期号：2
页码：35-51
语种：English
出版社：EduSoft publishing
摘要：Assembler Encoding is a neuro-evolutionary method in which a neural network is represented in the form of a simple program called Assembler Encoding Program. The task of the program is to create the so-called Network Definition Matrix which maintains all the information necessary to construct the network. To generate Assembler Encoding Programs and the subsequent neural networks evolutionary techniques are used. The performance of Assembler Encoding strongly depends on operations used in Assembler Encoding Programs. To select the most effective operations, experiments in the optimization and the predator-prey problem were carried out. In the experiments, Assembler Encoding Programs equipped with different types of operations were tested. The results of the tests are presented at the end of the paper.
关键词：artificial neural networks, assembler encoding program
