PhD Thesis

My PhD thesis is on « Collaborative Interactive Machine Teaching ». I am being supervised by Jules Françoise, Baptiste Caramiaux, and Michèle Gouiffès. This thesis is based on three fundamental elements. First, we are interested in the diverse roles that humans can take while interacting with a ML model. Second, our emphasis is on more democratized, participatory, and fair ML systems. Third, we are about to foster collaboration. In my last project we studied how a group of teachers can collaborate to teach a concept to an image classifier. If you are interested in the last sentece, you can read its paper. This project somehow demonstrated to us the new horizons of Fairness in ML models. Thus, we got curious to study another role a human can play in this interaction: the role of an auditor to discover, raise awareness of, and fix algorithmic harms. In a nutshell, my main interests (so far) are Interactive Machine Learning, Machine Teaching, Participatory AI, and End-User Algorithm Auditing.