A niching evolutionary algorithm with adaptive negative correlation learning for neural network ensemble

作者：Sheng, Weiguo^*; Shan, Pengxiao; Chen, Shengyong; Liu, Yurong; Alsaadi, Fuad E.

来源：Neurocomputing, 2017, 247: 173-182.

DOI：10.1016/j.neucom.2017.03.055

• Neural network ensemble
• Evolutionary algorithm
• Negative correlation learning
• Adaptation strategy
• Diversity measure

This paper proposes a niching evolutionary algorithm with adaptive negative correlation learning, denoted as NEAJNCL, for training the neural network ensemble. In the proposed NEA_ANCL, an adaptive negative correlation learning, in which the penalty coefficient A is set to dynamically change during training, has been developed. The adaptation strategy is based on a novel population diversity measure with the purpose of appropriately controlling the trade-off between the diversity and accuracy in the ensemble. Further, a modified dynamical fitness sharing method is applied to preserve the diversity of population during training. The proposed NEA_ANCL has been evaluated on a number of benchmark problems and compared with related ensemble learning algorithms. The results show that our method can be used to design a satisfactory NN ensemble and outperform related works.

• 出版日期2017-7-19
• 单位扬州大学; 天津理工大学; 杭州师范大学; 浙江工业大学