IEEE TRANSACTIONS ON NEURAL NETWORKS 1 A New Evolutionary System for Evolving Arti cialNeural

| This paper presents a new evolutionary system, i.e., EP-Net, for evolving artiicial neural networks (ANNs). The evolutionary algorithm used in EPNet is based on Fogel's evolutionary programming (EP) 1], 2], 3]. Unlike most previous studies on evolving ANNs, this paper puts its emphasis on evolving ANN's behaviours. This is one of the primary reasons why EP is adopted. Five mutation operators proposed in EPNet reeect such an emphasis on evolving behaviours. Close behavioural links between parents and their oospring are maintained by various mutations, such as partial training and node splitting. EPNet evolves ANN's architectures and connection weights (including biases 1) simultaneously in order to reduce the noise in tness evaluation. The parsimony of evolved ANNs is encouraged by preferring node/connection deletion to addition. EPNet has been tested on a number of benchmark problems in machine learning and ANNs, such as the parity problem, the medical diagnosis problems (breast cancer, diabetes, heart disease, and thyroid), the Australian credit card assessment problem , and the Mackey-Glass time series prediction problem. The experimental results show that EPNet can produce very compact ANNs with good generalisation ability in comparison with other algorithms.