Classifying Facial Expression with Radial Basis Function Netowrks , using Gradient Descent and K-means
暂无分享,去创建一个
This paper compares methods of training radial basis function networks. We found RBF networks initialized by supervised clustering perform better than networks initialized by unsupervised clustering and improved with gradient descent. Gradient descent did not significantly improve the networks initialized by supervised clustering.
[1] Heekuck Oh,et al. Neural Networks for Pattern Recognition , 1993, Adv. Comput..
[2] David M. Mount,et al. The analysis of a simple k-means clustering algorithm , 2000, SCG '00.
[3] Eric R. Ziegel,et al. The Elements of Statistical Learning , 2003, Technometrics.