Postdoc
April 2024-Current
IT University of Copenhagen
Copenhagen, Denmark
Robotics, Evolution, and Art Lab, under the supervision of Sebastian Rizi
Large Language Model for assitive AI in games
Publications:
Harnessing Language for Coordination: A Framework and Benchmark for LLM-Driven Multi-Agent Control
Adversarial Coevolutionary Illumination with Generational Adversarial MAP-Elites (under review)
Course manager:
2025-2026: Artificial Intelligence for Games and Simulations (15 ECTS), ITU, Copenhagen, Denmark.
PhD
September 2020-March 2024
Université de Lorraine, Inria Nancy-Grand Est, LORIA
Nancy, France
Larsen Team, under the supervision of Jean-Baptiste Mouret
Multi-task optimization and its application to robotics: first solve, then generalize
(manuscript in french)
Publications:
First do not fall: learning to exploit a wall with a damaged humanoid robot
Parametric-Task MAP-Elites
Multi-Task Multi-Behavior MAP-Elites
Teaching assistant:
2022-2023: Algorithmic, Functional Programming, and Digital Tools (64 hours) at Université de Lorraine, FST, Nancy, France.
2021-2022: Algorithmic, Functional Programming, System, and Digital Tools (64 hours) at Université de Lorraine, FST, Nancy, France.
2020-2021: Algorithmic, Functional Programming, Computer Graphics, Introduction to AI, and Advanced Methodology (64 hours) at Université de Lorraine, FST, Nancy, France.
Research Internship
January 2020-June 2020
Edinburgh Centre for Robotics, University of Edinburgh
Edinburgh, United Kingdom
Advanced Robotics Lab, under the supervision of Zhibin Li
Meta-learning for adaptive locomotion of a quadruped robot
Publication: Meta-Learning for Fast Adaptive Locomotion with Uncertainties in Environments and Robot Dynamics
(preprint available on arXiv)
September 2019-December 2019
INRIA Nancy Grand-Est
Villers-lès-Nancy, France
Team LARSEN, under the supervision of Jean-Baptiste Mouret
Meta-learning for adaptive locomotion of a quadruped robot
Report: Meta-learning for adaptive locomotion of a quadruped robot
Publication: Fast Online Adaptation in Robotics through Meta-Learning Embeddings of Simulated Priors (preprint available on arXiv)
February 2019-June 2019
INRIA Bordeaux Sud-Ouest
Talence, France
Team FLOWER, under the supervision of Pierre-Yves Oudeyer and Cédric Colas
Intrinsically Motivated Goal Exploration with Deep Model-Based Reinforcement Learning
Report: Intrinsically Motivated Goal Exploration with Deep Model-Based Reinforcement Learning
May 2018-August 2018
Center for Information and Neural Networks (CiNet)
National Institute of Information and Communication Technology (NICT)
Osaka, Japan,
Under the supervision of Yukie Nagai and Anja Philippsen
Report: Characterizing individual behaviors by using Recurrent Neural Network
Poster: Predictive Coding Account for the Development of Representational Drawing, 2019 SRCD Biennial Meeting
May 2017-July 2017
INRIA Bordeaux Sud-Ouest
Talence, France
Team FLOWER, under the supervision of Pierre-Yves Oudeyer and Sébastien Forrestier
Report (in Frensh): Comparative study of intrinsically motivated goal exploration algorithms