What this book covers

Chapter 1, Machine Learning for Cybersecurity, covers the fundamental techniques of machine learning for cybersecurity.

Chapter 2, Machine Learning-Based Malware Detection, shows how to perform static and dynamic analysis on samples. You will also learn how to tackle important machine learning challenges that occur in the domain of cybersecurity, such as class imbalance and false positive rate (FPR) constraints.

Chapter 3, Advanced Malware Detection, covers more advanced concepts for malware analysis. We will also discuss how to approach obfuscated and packed malware, how to scale up the collection of N-gram features, and how to use deep learning to detect and even create malware.

Chapter 4, Machine Learning for Social Engineering, explains how to build a Twitter spear-phishing bot using machine learning. You'll also learn how to use deep learning to have a recording of a target saying whatever you want them to say. The chapter also runs through a lie detection cycle and shows you how to train a Recurrent Neural Network (RNN) so that it is able to generate new reviews, similar to the ones in the training dataset.

Chapter 5, Penetration Testing Using Machine Learning, covers a wide selection of machine learning technologies for penetration testing and security countermeasures. It also covers more specialized topics, such as deanonymizing Tor traffic, recognizing unauthorized access via keystroke dynamics, and detecting malicious URLs.

Chapter 6, Automatic Intrusion Detection, looks at designing and implementing several intrusion detection systems using machine learning. It also addresses the example-dependent, cost-sensitive, radically-imbalanced, challenging problem of credit card fraud.

Chapter 7, Securing and Attacking Data with Machine Learning, covers recipes for employing machine learning to secure and attack data. It also covers an application of ML for hardware security by attacking physically unclonable functions (PUFs) using AI.

Chapter 8, Secure and Private AI, explains how to use a federated learning model using the TensorFlow Federated framework. It also includes a walk-through of the basics of encrypted computation and shows how to implement and train a differentially private deep neural network for MNIST using Keras and TensorFlow Privacy.

Appendix offers you a guide to creating infrastructure to handle the challenges of machine learning on cybersecurity data. This chapter also provides a guide to using virtual Python environments, which allow you to seamlessly work on different Python projects while avoiding package conflicts.