About Me

I work on representation learning for human behaviour. I am interested in developing models that better capture the inherent structure of behavioural data, including its temporal dynamics, multimodal nature, and discrete patterns. To this end, I design deep learning architectures that learn rich and structured representations of behaviour, which can be used not only for classification but also as tools for gaining deeper insights into human behaviour.

I am currently completing my Ph.D. at the Institut des Systèmes Intelligents et de Robotique (ISIR), Sorbonne Université, under the supervision of Mohamed Chetouani (ISIR) and David Cohen (ISIR, AP-HP), with my defence planned for September 2026. My doctoral research focused on representation learning for human behavioural wearable data. Funded through the EP PerMed TECH-TOYS project, my Ph.D. work explored machine learning approaches for the early detection of neurodevelopmental disorders, particularly cerebral palsy. I developed models that leveraged data collected through a sensorised play environment, enabling accessible and naturalistic assessment of children’s behaviour in home settings. I also worked on stress prediction from wearable data.

In summer 2024, I was a visiting Ph.D. student at the Cognitive Developmental Robotics Lab (Nagai Lab), IRCN, University of Tokyo, where I conducted research on learning representations of interpersonal synchrony.

Before this, I received my Master’s degree in Intelligent Systems from Sorbonne University, where I also completed my Bachelor’s degree in Electronics and Electrical Engineering.

My current research interests include:

• Representation Learning
• Infant development, Language and social skills acquisition
• Social Signal Processing