When you are shopping online or surfing movies, you may wonder how the products you may also like or movies that may also interest you works. In this chapter, we will explain the algorithm behind the scene, called the restricted boltzmann machine (RBM). We will start with reviewing RBMs and their evolution path. We will then dig deeper into the logic behind them and implement RBMs in TensorFlow. We will also apply them to build a movie recommender. Beyond a shallow architecture, we will move on with a stacked version of RBMs called deep belief networks (DBNs) and use it to classify images, of course, with our implementation in TensorFlow from scratch.
RBMs find a latent representation of the input by attempting to reconstruct the input data. In this chapter, we will also discuss the autoencoders as another...
