Deep Belief Nets - Ep. 7 (Deep Learning SIMPLIFIED)

An RBM can extract features and reconstruct input data, but it still lacks the ability to combat the vanishing gradient. However, through a clever combination of several stacked RBMs and a classifier, you can form a neural net that can solve the problem. This net is known as a Deep Belief Network. Deep Learning TV on Facebook: Twitter: The Deep Belief Network, or DBN, was also conceived by Geoff Hinton. These powerful nets are bel