Title :
Designing neural networks ensembles based on the evolutionary programming
Author :
Liu, Fang ; Li, Ren-Hou ; Mei, Shi-cwn
Author_Institution :
Syst. Eng. Inst., Xi´´an Jiaotong Univ., China
Abstract :
An evolutionary programming is proposed in this paper to automatically design neural networks (NNs) ensembles. Based on negative correlation learning, different individual NNs in the ensemble can learn to subdivide the task and thereby solve it more efficiently and elegantly. At the same time, different individual NNs are always to find the best collaboration connection during the evolutionary process. In addition, the architecture of each NN in the ensemble and the size of the ensemble need not to be predefined. The simulation results show that the proposed method in this paper is valid.
Keywords :
correlation methods; evolutionary computation; learning (artificial intelligence); neural nets; ensemble learning system; evolutionary programming; negative correlation learning; neural networks; Algorithm design and analysis; Automatic programming; Cybernetics; Design engineering; Genetic programming; Learning systems; Machine learning; Neural networks; Robustness; Training data;
Conference_Titel :
Machine Learning and Cybernetics, 2003 International Conference on
Print_ISBN :
0-7803-8131-9
DOI :
10.1109/ICMLC.2003.1259724
