3 - Fairness, explainability, privacy, and robustness for trustworthy algorithmic decision-making

Abstract

With the rapid increase in the use and deployment of machine learning (ML) systems in the world, concomitant concerns on the ethical implications of their downstream effect have surfaced in recent years. Responding to this challenge, the field of trustworthy ML has grown rapidly and resulted in a large body of methods and algorithms that embody desirable qualities such as fairness, transparency, privacy, and robustness. In this chapter, we survey the current landscape of trustworthy ML methods, introduce fundamental concepts, and summarize research directions. To bridge the gap between theory and practice, we provide implementation details of each category of methods that are currently available publicly.