Chapter 6. Deep learning and neural networks

 

This chapter covers

  • Neural network basics
  • An introduction to deep learning
  • Digit recognition using restricted Boltzmann machines

There is much discussion about deep learning at the moment, and it’s widely seen to be the next big advance in machine learning and artificial intelligence. In this chapter, we’d like to cut through the rhetoric to provide you with the facts. At the end of this chapter, you should understand the basic building block of any deep learning network, the perceptron, and understand how these fit together in a deep network. Neural networks heralded the introduction of the perceptron, so we’ll discuss these before exploring deeper, more expressive networks. These deeper networks come with significant challenges in representation and training, so we need to ensure a good foundational knowledge before leaping in.

6.1. An intuitive approach to deep learning

6.2. Neural networks

6.3. The perceptron

6.4. Multilayer perceptrons

6.5. Going deeper: from multilayer neural networks to deep learning

6.6. Summary