I am a Research Manager at Microsoft Research Montréal. I completed a Ph.D. in computer science at Université de Lorraine in France in 2016. My Ph.D. was a joint project between Université de Lorraine and Orange Labs, the research and development branch of Orange, a French telecommunication company. My research focused on learning the parameters of reinforcement learning for dialogue systems. I proposed methods to learn a reward function from a set of dialogues which were rated by humans. I also worked on learning an interpretable representation of the state space.
I joined Maluuba, a Canadian startup, in 2016 as a research scientist and I worked on user simulation and reinforcement learning for dialogue systems. I then became a team lead and led the data collection for the Frames dataset: a goal-oriented dialogue dataset. Microsoft acquired Maluuba in 2017 and I have been ever since a research manager leading a team focused on dialogue research. The team works on advancing the state-of-the-art as well as improving Microsoft products with dialogue capabilities.
My main interests are currently: the evaluation of dialogue systems, offline learning on human-human dialogues, and multimodal dialogue.