Tied to supervised machine learning, the popularity of neural networks has skyrocketed in the last few years. A major factor for this was the open source release of TensorFlow in 2017, and of simplified APIs for it, such as Keras. But the concepts underlying neural networks are as old as computers. However, they were not at the center of the dominant branch of computer developments and, for some time, were even discredited. Interestingly, they embody radically different ways of thinking about computational problems, and even about computers. The next few chapters show the several new conceptual tools brought about by neural networks, with and without learning.
The examples here use Keras with the TensorFlow backend. Even ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access