Machine Learning Security Principles

Table of Contents

Preface

Part 1 – Securing a Machine Learning System

1

Defining Machine Learning Security

Building a picture of ML

Why is ML important?

Identifying the ML security domain

Distinguishing between supervised and unsupervised

Using ML from development to production

Adding security to ML

Defining the human element

Compromising the integrity and availability of ML models

Describing the types of attacks against ML

Considering what ML security can achieve

Setting up for the book

What do you need to know?

Considering the programming setup

Summary

2

Mitigating Risk at Training by Validating and Maintaining Datasets

Technical requirements

Defining dataset threats

Learning about the kinds of database threats

Considering dataset threat sources ...