Abstract
A Hybrid Genetic Programming (HGP) algorithm is proposed for optimal approximation of high order and sparse linear systems. With the intrinsic property of linear systems in mind, an individual in HGP is designed as an organization that consists of two cells. The nodes of the cells include a function and a terminal. All GP operators are designed based on organizations. In the experiments, three kinds of linear system approximation problems, namely stable, unstable, and high order and sparse linear systems, are used to test the performance of HGP. The experimental results show that HGP obtained a good performance in solving high order and sparse linear systems. © 2008 Springer Berlin Heidelberg.
Cite
CITATION STYLE
Liu, J., Fu, W., & Zhong, W. (2008). Hybrid genetic programming for optimal approximation of high order and sparse linear systems. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 5361 LNAI, pp. 462–472). https://doi.org/10.1007/978-3-540-89694-4_47
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
