AIRC Team

Giulia is currently undergoing a PhD research project, funded by the Technological University Dublin, in eXplainable Artificial Intelligence (XAI). XAI is a new branch of Artificial Intelligence that aims at developing novel methods capable of exposing and explaining the logic followed by data-driven machine-learned models. The main goal of this project is to design, develop and evaluate a multi-layer argument-based framework for explainability. The arguments will be based on rulesets automatically extracted from complex models, perceived as black-boxes by users, by a post-hoc, model-agnostic rule-extraction XAI method. The addition of an argument-based layer to rule-based explanations represents an automatic way to summarise the inferential process of an underlying model, thus enhancing its degree of explainability, without any human intervention. These explanations will be compared to another format of rule-based explanations, namely decision trees, which are considered naturally intelligible and transparent. Both explanations will be also presented to the participants of a human study to collect their qualitative perceptions. Previously, Giulia earned a Masters in Applied Mathematics, a Diploma in Actuarial Sciences and 10+ years experience in complex data modelling in the life insurance and exposure science worlds, liaising with large multinational companies and industry associations.