12INTRODUCTION TO CONVOLUTIONAL NEURAL NETWORKS
In this chapter, we’ll introduce a new and potent approach to dealing with multidimensional information. In particular, we’ll work through the theory and high-level operation of convolutional neural networks (CNNs), a cornerstone of modern deep learning.
We’ll begin by presenting the motivations behind the development of CNNs. Convolutions are the heart of CNNs, so they’ll come next. We’ll discuss them in some detail, in particular how they’re used by the CNN. We’ll then introduce a basic CNN and work through its anatomy. We’ll use this basic CNN architecture for the remainder of the chapter. After ...
Get Practical Deep Learning now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
