Abstract
To address the rising concern that algorithmic decision-making may reinforce discriminatory biases, researchers have proposed many notions of fairness and corresponding mathematical formalizations. Each of these notions is often presented as a one-size-fits-all, absolute condition; however, in reality, the practical and ethical trade-offs are unavoidable and more complex. We introduce a new approach that considers fairness—not as a binary, absolute mathematical condition—but rather, as a relational notion in comparison to alternative decisionmaking processes. Using US mortgage lending as an example use case, we discuss the ethical foundations of each definition of fairness and demonstrate that our proposed methodology more closely captures the ethical trade-offs of the decision-maker, as well as forcing a more explicit representation of which values and objectives are prioritised.
Highlights
Algorithms are increasingly being used to make important decisions to improve efficiency, reduce costs, and enhance personalisation of products and services
The data used are collected under the Home Mortgage Disclosure Act (HMDA) from all lenders above a size threshold that are active in metropolitan areas
The one-size-fits-all, binary, and axiomatic approaches to fairness are insufficient in addressing the complexities of racial discrimination in mortgage lending
Summary
Algorithms are increasingly being used to make important decisions to improve efficiency, reduce costs, and enhance personalisation of products and services. Despite these opportunities, the hesitation around implementing algorithms can be attributed to the risk that the algorithms may systematically reinforce past discriminatory practices, favouring some groups over others on the basis of race and gender and thereby exacerbating societal inequalities. A decision based on an algorithm is less transparent or explainable than a rules-based process (e.g. if X, Y). Whereas a human decision-maker with the same cognitive biases may vary his or her decisions, an algorithmic decision based on a bias is capable of perpetuating discrimination at-scale
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
