Fig. 1From: Bayesian neural networks with variable selection for prediction of genotypic valuesNetSparse Schematic neural network representation of NetSparse (9). The input \(\mathbf {x}\) to the neural network is on the left, the output g is on the right. \(\mathbf {s}\) is the variable selection vector, \(\mathbf {W}\) and \(\mathbf {w}\) are the weights, \(\mathbf {b}^h\) and \(b^o\) are the biases. At the third layer of nodes, \(\tanh\) is applied to the sum of the incoming valuesBack to article page