Platform and Model Design for Responsible AI

7

Fairness Notions and Fair Data Generation

In this chapter, we will first set an outline of how fairness has become important in the world of predictive modeling by providing examples of different challenges faced in society. We will then go deep into the taxonomies and types of fairness to present a detailed description of the terms involved. Here, we will understand the importance of the defined metrics by citing and substantiating open source tools that help evaluate the metrics. Then, we will further emphasize the importance of the quality of data as biased datasets can introduce hidden bias in ML models. In this context, this chapter discusses different synthetic data generation techniques that are available and how they can be effective ...