The next Comète seminar will take place on Friday, November 29th at 2pm in Salle Emmy Noether (LIX - Alan Turing building, École Polytechnique). Karima Makhlouf (Université du Quebec à Montréal (UQAM), Canada) and Sami Zhioua (Higher Colleges of Technology, Dubai) will talk about Intuitive Introduction to Fairness and Privacy in Machine Learning.

Abstract: Machine Learning (ML) algorithms have penetrated every aspect of our daily life. From movie recommendations to driving autonomous cars, ML outperforms humans in algorithmic decision-making as they can take into account orders of magnitude more factors than people can. The dark side, however, is that ML algorithms are often vulnerable to biases that render their decisions “unfair” towards certain individuals and minorities. As ML is increasingly used in high-impact/life-changing decision making such as job hiring, University admission, and predicting whether released people from jail will re-offend, fairness becomes critically important. In this presentation, we explore the problem of fairness and privacy in machine learning. In particular, we discuss the different and often conflicting definitions of fairness, the origins of bias, and the currently existing approaches to ensure fairness in ML algorithms. The presentation is intended to be an intuitive introduction to the field.