Scientific contributions

• Intersectional metrics for measuring data and models bias index at local and global federated learning settings (protocol and mathematical model). 
• Intersectional approaches, aware of diversity and inclusion perspectives for designing data science pipelines (mathematical model). 
• Certification and negotiation-based approach for assessing intersectional bias levels in federated learning (model and protocol). 
• Experimentations (experiment setting + software on Github + statistics + paper).

Dissemination and consolidation of the collaboration

• Publications (2/year) submitted, for example, in EDBT, CAISE, and Springer Cloud Computing Journal, depending on the status of the results.
• Master student reports with paper versions on Arxiv.
• By the end of the first semester of 2024, with evidence of strong collaboration, we will approach either a European, French Agence Nationale de la Recherche (ANR) or a binational collaboration call (e.g., University Franco-Italienne UFI) to pursue collaborative proposals.