Restricted Boltzmann Machine
Restricted Boltzmann MachineIt is a random neural network model with a two-layer structure, stacked connections and no autonomous feedback. Its characteristics are full connection within the layer and no connection outside the layer. It was proposed by Geoff Hinton and others from the University of Toronto and can be used for dimensionality reduction, classification, regression, collaborative filtering, feature learning and subject modeling algorithms.
The upper layer of neurons in the restricted Boltzmann machine constitutes the hidden layer, and the h vector represents the value of the neurons in the hidden layer; the lower layer of neurons constitutes the visible layer, and the v vector represents the value of the neurons in the visible layer.
Unlike the Boltzmann machine, the restricted Boltzmann machine has no connections within the layer, so it can be said to be restricted, that is, a simplified Boltzmann machine model.