**Boltzmann Machines** are bidirectionally connected networks of stochastic processing units, i.e. units that carry out randomly determined processes.

**Read the full article for complete details and intuition behind Restricted Boltzmann Machines**and join the discussion now

A Boltzmann Machine can be used to

**learn important aspects of an unknown probability distribution based on samples from the distribution**. Generally, this learning problem is quite difficult and time consuming. However, the learning problem can be simplified by introducing restrictions on a Boltzmann Machine, hence why, it is called a

**Restricted Boltzmann Machine**.

**Restricted Boltzmann Machines**has found applications in

**Quantum Computing**and is commonly used in dimensionality reduction, classification, feature learning, topic modelling , recommendation systems and music generation.

This is a companion discussion topic for the original entry at http://iq.opengenus.org/restricted-boltzmann-machine/