id	author	title	date	pages	extension	mime	words	sentence	flesch	summary	cache	txt
ejpam-1375	Akbilgic, Oguz; Bozdogan, Hamparsum	Predictive Subset Selection using Regression Trees and RBF Neural Networks Hybridized with the Genetic Algorithm	2011	20	.pdf	application/pdf	7206	393	63	Least Squares Estimation Given a network (or model) in (1) consisting of m RBFs with centers ¦ c j ©m j=1 and radii¦ r j ©m j=1 and a training set with p patterns, �� x i, yi � p i=1 , the optimal network weights can be found by minimizing the sum of squared errors: SSE = p∑ i=1 � f (x i)− yi �2 (8) and is given by ŵ = � H ′ H �−1 H ′ y (9) the so called normal equation. Hi j = h j(x i), and y = � y1, y2, . . .	cache/ejpam-1375.pdf	txt/ejpam-1375.txt