A significant challenge in detecting and mitigating bias is creating a mindset amongst AI developers to address unfairness. The current literature on fairness is broad, and the learning curve to distinguish where to use existing metrics and techniques for bias detection or mitigation is difficult. This survey systematises the state-of-the-art about distinct notions of fairness and relative techniques for bias mitigation according to the AI lifecycle. Gaps and challenges identified during the development of this work are also discussed.
Assessing and Enforcing Fairness in the AI Lifecycle / Calegari R.; Castane G.G.; Milano M.; O'Sullivan B.. - In: IJCAI. - ISSN 1045-0823. - ELETTRONICO. - 2023-:(2023), pp. 6554-6562. (Intervento presentato al convegno 32nd International Joint Conference on Artificial Intelligence, IJCAI 2023 tenutosi a Macau, China nel August 2023) [10.24963/ijcai.2023/735].
Assessing and Enforcing Fairness in the AI Lifecycle
Calegari R.;Milano M.;
2023
Abstract
A significant challenge in detecting and mitigating bias is creating a mindset amongst AI developers to address unfairness. The current literature on fairness is broad, and the learning curve to distinguish where to use existing metrics and techniques for bias detection or mitigation is difficult. This survey systematises the state-of-the-art about distinct notions of fairness and relative techniques for bias mitigation according to the AI lifecycle. Gaps and challenges identified during the development of this work are also discussed.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.