Behavioural Computing for Artificial Intelligence
Human-computer interfaces using artificial intelligence for real-life applications are nowadays ubiquitous and utilised by a large diversity of population. However, systems lack the ability to understand humans beyond the expressed message, and thus fail at providing personalised services. The automatic sensing of human behaviours can provide meaningful information of the user, and allow artificial intelligence to adaptively fit his/her peculiarities. The automatic sensing of human behaviours expressed in real-life conditions is yet challenging. This talk will present recent advances in the field along with current limitations, and draw perspectives for the next generation of artificial intelligence.
Fabien Ringeval received the Ph.D. degree for his researches on the automatic recognition of acted and spontaneous emotions from speech in 2011, from the Université Pierre et Marie Curie (UPMC), Paris, France. He is Assistant Professor at the Laboratoire d'Informatique de Grenoble (LIG), CNRS, Université Grenoble Alpes, France, since 2016. Dr. Ringeval is also a senior researcher at audEERING GmbH since 2016. His research interests concern digital signal processing and machine learning, with applications on the automatic recognition of paralinguistic information (e.g., emotions, social and atypical behaviours) from multimodal data, at the cross-road of computer sciences and human behaviour understanding.