In interactive machine learning, humans and machine learning algorithms collaborate to achieve tasks mediated by interactive visual interfaces. In traditional interactive machine learning, models are often developed with single users in mind. However, real world applications often require the collaboration of different types and levels of expertise to help reliably develop and assess the results of machine learning models. The goal of this PhD is to develop a collaborative interactive machine learning (CoCo-IML) framework, and a prototype system that implements it, in order to: (i) on the one hand support the dialogue between different types of domain experts and modelers, and (ii) on the other hand, support the dialogue between the experts and the machine learning algorithm. The proposed work will be demonstrated through a real world use-case application from the agronomy and food technology domains, following a participatory design methodology. The evaluation of the framework and prototype system will take into account various criteria including the interpretability and user trust of the model, as well as the quality of collective decision making under uncertainty.
This PhD topic is participating to the Université Paris-Saclay EU COFUND DeMythif.AI program : https://www.dataia.eu/actualites/cofund-demythifai-appel-sujets-de-these. It is reserved to international students who have spent less than 12 months in France in the last 3 years. The candidates will be evaluated by a jury who will select 15 PhD to start in fall 2024. The successful candidates will be fully funded for 3 years, have access to specific scientific and non-scientific training, and be fully part of the Université Paris-Saclay AI community.