Scientific Course: Monday, June 21
"Mathematical models for fairness in machine learning" by Jean-Michel Loubes and Laurent Risser (Université Toulouse Paul Sabatier & Artificial and Natural Intelligence Toulouse Institute (ANITI), France) on Monday June 21, 2021.
10:00 - 12:00 and 14:30 - 16:30
The study of bias in Machine Learning is nowadays an important issue due to the development of AI's driven algorithm in our everyday lives. We propose in this short course to study how to clarify certain notions of the study of bias in machine learning. We will see how to model fairness for mathematical models: group fairness but also individual fairness. We will review the major ways to measure and then ensure fairness in supervised learning. We will focus on the consequences of the new European Artificial Intelligence Act (04/2021) on the regulation in the use of ML algorithms. A python notebook will be made available to test some of the methods presented here (the presentation will be done by Laurent Risser).