Deep Belief Nets in C++ and CUDA C: Volume 1: Restricted Boltzmann Machines and Supervised Feedforward Networks

© Timothy Masters 2018

Timothy MastersDeep Belief Nets in C++ and CUDA C: Volume 1https://doi.org/10.1007/978-1-4842-3591-1_2

2. Supervised Feedforward Networks

Timothy Masters¹ 

(1)

Ithaca, New York, USA

 

Deep belief nets are generally trained in stages. First, one or more (usually more) layers are trained with unsupervised algorithms. Rather than seeking to learn class memberships or predicted values, the model simply tries to find consistent patterns within the independent variables. Only after such patterns have been found does training switch to supervised mode. However, because supervised training algorithms are easier to understand than the usual unsupervised algorithms, we will begin this study of deep belief nets with supervised training. ...