Bio
Madeleine Waller is a PhD researcher at King's College London in the UKRI Safe and Trusted AI Centre for Doctoral Training. Her research focuses on the fairness of automated decision-making systems. Before her PhD studies, Madeleine received a First Class (Honours) Computer Science BSc at King's College London. Madeleine's main interest is in AI policy, specifically around preventing the use of discriminatory AI systems.
Research interests
Madeleine's research develops methods that detect, explain and mitigate unfairness in binary classification decision-making systems that are based on tabular data. Specifically, she uses approaches from computational argumentation to construct graphs representing the individual fairness of a decision made by a black-box system and uses these graphs to explain the decisions. Feedback from a user is then used to mitigate the potential bias and potentially change the decision made by the system. Her research links the fields of AI fairness, transparency and explainability.