Beginning Anomaly Detection Using

generative network 

(meaning it’s capable of generating its own data based on what it has learned by fitting 

on a data set). In Boltzmann machines, the visible nodes are what we can interact with; 

we can’t interact with the hidden nodes. One more distinction to make is that there is no 

training process; the nodes learn to model the data set as best as they can on their own, 

making the Boltzmann machine an 

unsupervised deep learning model.

However, Boltzmann machines aren’t necessarily that practical, and they suffer from 

problems when the network is scaled up in size. Specific derivations of the Boltzmann 

machine such as