Fairness is a concept central to ethical decision-making, can be a complex notion to define and measure, particularly because it can be subjective and context-dependent. In essence, fairness in data science and AI refers to the idea that decisions made by these systems should be equitable and just, not favoring one group over another based on characteristics like gender, race or ethnicity.

While it’s true that reasonable people can have differing opinions on what constitutes an ethical decision, there should be consensus on what needs to be measured and incorporated into decision-making processes. As Simon Winchester points out: “All life depends to some extent on measurement, and in the very earliest days of social organization a clear indication of advancement and sophistication was the degree to which systems of measurement had been established, codified, agreed to and employed”(1). In this context, tools like the AEQUITAS toolkit can be invaluable. This toolkit provides a robust framework to measure the biases in any decisions that are made, offering a practical way to operationalize and quantify fairness.

Figure 1: Source: http://www.datasciencepublicpolicy.org/our-work/tools-guides/aequitas/

Aequitas is an open-source bias and fairness audit toolkit designed to aid data scientists and policymakers in evaluating biases in machine learning models across various population sub-groups. Available as a Python library, command line interface, or web application, Aequitas is versatile and user-friendly. Recognizing that fairness is context-dependent, Aequitas offers comprehensive guidance for its application in public policy, emphasizing the importance of understanding potential interventions and their implications. The toolkit can audit models at various stages: before model selection, before operationalization, or during A/B testing of revised algorithms. Aequitas primarily serves two user groups:

Data Scientists and AI Researchers, who employ the toolkit during the model building and selection phase to measure and compare biases in different models.
Policymakers, who utilize Aequitas to understand and address biases in AI systems before implementing them in policy decisions. They also need to periodically reassess models to monitor fairness degradation over time.

In essence, Aequitas ensures continuous evaluation of biases in machine learning models, both internally by data scientists during model development and externally by policymakers before and after model deployment.

Figure 2: Source: http://www.datasciencepublicpolicy.org/our-work/tools-guides/aequitas/

Figure 2 positions Aequitas within the machine learning workflow, indicating when and by whom audits should be conducted. Aequitas aims to streamline the process of identifying model biases. This toolkit, tailored for both data scientists and decision-makers, ensures bias and fairness are addressed throughout the model's lifecycle, including model selection, deployment decisions, retraining considerations, and data collection needs.

Reference:

1. Winchester S. The perfectionists: how precision engineers created the modern world: HarperLuxe; 2018.