Fairness: from the ethical principle to the practice of Machine Learning development as an ongoing agreement with stakeholders
This work tackles the problem of integrating fairness into ML development for practitioners and stakeholders, offering a structured approach to manage bias as an incremental improvement in ethical AI practices.
The paper addresses the inherent inability to completely eliminate bias in machine learning by proposing an end-to-end methodology that translates ethical fairness principles into practice through ongoing stakeholder agreements, aiming to challenge power dynamics and guide teams in identifying, mitigating, and monitoring bias throughout development.
This paper clarifies why bias cannot be completely mitigated in Machine Learning (ML) and proposes an end-to-end methodology to translate the ethical principle of justice and fairness into the practice of ML development as an ongoing agreement with stakeholders. The pro-ethical iterative process presented in the paper aims to challenge asymmetric power dynamics in the fairness decision making within ML design and support ML development teams to identify, mitigate and monitor bias at each step of ML systems development. The process also provides guidance on how to explain the always imperfect trade-offs in terms of bias to users.