Strengthening Deep Neural Networks

Chapter 3. Deep Neural Network (DNN) Fundamentals

In this chapter, we’ll explore the core concepts behind DNN models, the category of machine learned models usually used for image and audio processing. Comprehending these basic ideas now will help you understand adversarial examples in greater depth later in the book. Following this basic introduction, Chapter 4 will then explore models for understanding complex images, audio, and video. The two chapters will provide sufficient background for the discussions on adversarial examples that follow but are not intended to provide a comprehensive introduction to deep learning.

If you are familiar with the principles of deep learning and neural networks, feel free to skip this chapter and Chapter 4. Conversely, if you are inclined to learn more than is required for this book, there are numerous excellent resources available to gain a better understanding of machine learning and neural networks. Links to a selection of online resources are included in this book’s GitHub repository.

At the end of this chapter, there are some snippets of Python code. As with all the code in this book, reading it is optional (you can understand adversarial examples without the code). If you are interested in the code, I encourage you to also download and experiment with some of the Jupyter notebooks provided in the associated GitHub repository.

Machine Learning

DNNs belong to the broader category of machine learning (ML); the capability for a machine to ...