Marc Hulcelle

I am a Post-doctoral researcher at the DiiP Institute affiliated to the Université Paris Cité. My research interests are at the crossroad of Machine Learning and Sociology, where I study social inequalities. I am working on two different projects. First, I study fairness methods under data drift constraints applied to the home property differences between US states. Second, I study the correlation between grammatical/pronominal gender and the existence of gender disparities in societies all over the world.

I conducted my PhD studies in the Signal, Statistics and Learning (S2A) team in the Image, Data and Signal (IDS) Department of the Laboratory Treatment and Communication of Information (LTCI) at Télécom Paris, Palaiseau, under the supervision of Chloé Clavel (Télécom Paris), Giovanna Varni (Télécom Paris), and Nicolas Rollet (i3 Lab, Télécom Paris). My work focused on the analysis of trust in Human-Robot Interaction (HRI). In particular, I created an annotation scheme named TURIN to study trust throughout the entire interaction, and built several multimodal computational models of trust by leveraging Interactionist Sociology theories. I aim to provide a better explanation of how trust builds and develops during an interaction with said models. I also explored the differences between currently existing Psychological models of trust in HRI that adopt a mentalist approach and Interactionist Sociology theories for trust Analysis as can be done in Conversation Analysis. Through these differences, I investigated how these two approaches can complement each other.

Before my PhD, I worked at Blue Frog Robotics as a robotics software engineer where I worked on developing both the AI of the Buddy robot (R&D on natural interactions, robot "personnality") and its software (architecture, remote-control application). During an academic exchange program at Tohoku University, I conducted research on obstacle detection using a time-of-flight camera for a moon exploration rover at the Space Robotics Lab.

I also hold an Engineering Degree from EMSE (France).

Publications

Journals

• “Qualification of a time-of-flight camera as a hazard detection and avoidance sensor for a moon exploration microrover”. K. Uno, L.-J. Burtz, M. Hulcelle, and K. Yoshida in Transactions of The Japan Society for Aeronautical and Space Sciences, Aerospace Technology Japan, vol. 16, no. 7, pp. 619–627, 2018.

Conferences

• "Comparing a mentalist and an interactionist approach for trust analysis in Human-Robot Interaction". M. Hulcelle, G. Varni, N. Rollet, and C. Clavel. In Proceedings of the 11th International Conference on Human-Agent Interaction (HAI '23). Association for Computing Machinery, New York, NY, USA, 273–280.
• "Leveraging Interactional Sociology for Trust Analysis in Multiparty Human-Robot Interaction". M. Hulcelle, L. Hemamou, G. Varni, N. Rollet, and C. Clavel. In Proceedings of the 11th International Conference on Human-Agent Interaction (HAI '23). Association for Computing Machinery, New York, NY, USA, 484–486
• “Computational Multimodal Models of Users' Interactional Trust in Multiparty Human-Robot Interaction”. M. Hulcelle, G. Varni, N. Rollet, and C. Clavel in 2022 12th Workshop on Human Behavior Understanding (ICPR), 2022.
• “Turin: A coding system for trust in human robot interaction”. M. Hulcelle, G. Varni, N. Rollet, and C. Clavel in 2021 9th International Conference on Affective Computing and Intelligent Interaction (ACII), 2021, pp. 1–8.
• "Exploring User's Trust Dynamics in Human-Robot Interaction". M. Hulcelle, G. Varni, N. Rollet, and C. Clavel in Top Ranked Abstracts from the 2021 Annual Meeting of the Society for Affective Science. Affec Sci (2021).

Thesis

• “Automatic analysis of trust over the course of a human-robot interaction using multimodal features and recurrent neural architectures”. M. Hulcelle

Academic duties

Contact