Indian Journal of Science and Technology
DOI: 10.17485/ijst/2016/v9iS1/110176
Year: 2016, Volume: 9, Issue: Special Issue 1, Pages: 1-6
Original Article
C. K. Ang* , M. I. Solihin and T. H. Tan
Faculty of Engineering, Technology and Built Environment, UCSI University, Kuala Lumpur, Malaysia; [email protected]
[email protected]
[email protected]
*Author for correspondence
Ang
Faculty of Engineering
Email: [email protected]
Back-Propagation Neural Network (BPNN) has been widely used in solving nonlinear problems. However, there are some limitations in using conventional BPNN especially for high order nonlinear problems. Dynamic Back-Propagation Neural Network (DBPNN) is proposed in this paper to improve the performance of conventional BPNN. Its adaptive learning ability is closer to human being learning behavior in comparing to conventional BPNN. Few simulations have been run to test the robustness of DBPNN and the results are compared to the conventional BPNN
Keywords: Artificial Intelligences, Neural Networks, Nonlinear Function
Subscribe now for latest articles and news.