Restricted Boltzmann Machines (RBM)
Restricted Boltzmann Machines RBM

RBMs are similar to Boltzmann Machine (BM) with restriction in connecting groups of neurons to another group(instead of Hopfield Networks (HN) like full connection) being their biggest difference which makes them more usable.

RBMs can be trained like FeedForward Neural Networks (FF or FFNN) with a twist: instead of passing data forward and then back-propagating, you forward pass the data and then backward pass the data (back to the first layer). After that you train with forward-and-back-propagation.


  • It's an unsupervised learning algorithm.
  • Composition: visible/input layer and the hidden layer without any intra-layer connections