Quentin Gallouédec

Ph.D. Student in Reinforcement Learning

Biography

I am a Ph.D. student working on Reinforcement Learning at LIRIS/École Centrale de Lyon, supervised by Associate Prof. Emmanuel Dellandréa. I focus on the design of robust reinforcement learning algorithms, especially for highly sparse reward environments. I am currently exploring efficient exploration with the Go-Explore paradigm and curiosity-based methods. My driving force is my fascination for the intertwining of cognitive science and artificial intelligence. Every day I ask myself this very basic question: how do we learn so much without even realizing it?

Feel free to contact me, I would be happy to share with you my work or anything else.

Interests

Artificial Intelligence
Reinforcement Learning
Hard-exploration Learning
Robotics Learning

Publications

Quentin Gallouédec (2024). Toward the Generalization of Reinforcement Learning.

PDF Cite Defense

Quentin Gallouédec, Edward Beeching, Clément Romac, Emmanuel Dellandréa (2024). Jack of All Trades, Master of Some, a Multi-Purpose Transformer Agent. arXiv preprint arXiv:2402.09844.

PDF Cite Paper Code Model

Quentin Gallouédec, Emmanuel Dellandréa (2022). Cell-Free Latent Go-Explore. Proceedings of the 40th International Conference on Machine Learning, PMLR 202:10571-10586, 2023.

PDF Cite Paper Code Poster

Quentin Gallouédec, Nicolas Cazin, Emmanuel Dellandréa, and Liming Chen (2021). panda-gym: Open-source goal-conditioned environments for robotic learning. Robot Learning Workshop: Self-Supervised and LifelongLearning at NeurIPS.

PDF Cite Paper Talk