Beginning Anomaly Detection Using

 Restricted Boltzmann Machine (RBM)

The 

RBM is similar to the Boltzmann machine in that it is an unsupervised, stochastic 

(probabilistic), generative deep learning model. However, a key difference is that 

the RBM is only comprised of two layers: the input layer and the hidden layer. Its 

architecture is similar to that of the artificial neural network model you explored in 

Chapter 

3

, with the RBM layers looking like the first two layers of an ANN. Because 

we place a restriction on the layers that none of the nodes within their own layer are 

to be interconnected, the model is termed as a 

restricted Boltzmann machine. More 

specifically, since each node outputs a binary value, we are dealing with a