Vincent Mousseau and Wassila Ouerdane (CentraleSupélec) ran a Seminar@SystemX on April 11, on the following topic: “Interactive Recommendation and Explanation for Multiple Criteria Decision Analysis”.
From the first expert systems to the recent recommender systems, decision-aiding has been a central concern in Artificial Intelligence. Very soon, it has become clear that providing recommendations was only part of the challenge: as decision-aiding tools become everyday more popular and more sophisticated, it is of utmost importance to develop their explanatory capabilities. Consequently, “Why should I trust you?” is a question every system should be prepared to answer – especially when the stakes are high. More precisely, there is a growing demand of institutions and citizens to make algorithmic decisions transparent and trustworthy. We interpret this requirement in the strong sense of “accountability”, when the recipient of the recommendation is able to defend it before other, skeptical, stakeholders of the decision. In this seminar, we will present how the requirements of explainable Artificial Intelligence shed a new light on multiple criteria decision aid, and in particular how interactive process by which a preference model is learned from the DM can be enhanced by explanation features.
In 1993, Vincent Mousseau obtained a PhD in Computer Science at the University of Paris-Dauphine, and went on to complete his HDR in 2003. Assistant Professor at the same University from 1994 to 2008, he has since served as Professor at CentraleSupélec in the LGI. Within this lab, he heads since 2008 the Decision Aid research team. Since 2013, he is the head of the axis ‘Complex System Engineering’ of the Doctoral School ‘Interfaces’ at “Université Paris Saclay”. His research is in the field of multiple criteria decision aiding. He also developed many collaborations on this field academic and socio economic partners. He is heavily involved in the Decision Deck European project as president of the consortium. He is Editor in Chief of the European Journal of Decision Process (EJDP).
Wassila Ouerdane is an assistant Professor in Computer Science within the LGI lab at CentraleSupélec, since September 2010. She obtained her PHD in 2009 at Université Paris Dauphine. Her research activities are accross two fields: Artificial Intelligence and Multiple Criteria Decision aiding. She investigates several questions among them: constructing explanation (argumentation) within a decision aiding process, dealing with the dynamic of decision models within a decision process. She is also interested by applying decision aiding methodologies within real context. Find out more.