Research articles
2023
[12] We won't get fooled again: when performance metric malfunction affects the landscape of hyperparameter optimization problems, René Traoré, Andrés Camero, Xiao Xiang Zhu, International Conference on Optimization and Learning, Springer, 2023.
[11] A Data-driven Approach to Neural Architecture Search Initialization, René Traoré, Andrés Camero, Xiao Xiang Zhu, Annals of Mathematics and Artificial Intelligence, Springer Nature, 2023.
2022
[10] HPO: we won’t get fooled again, René Traoré, Andrés Camero, Xiao Xiang Zhu, International Conference on Automated Machine Learning, Late breaking workshop 2022.
[9] Landscape of Neural Architecture Search across sensors: how much do they differ ? René Traoré, Andrés Camero, Xiao Xiang Zhu, Annals of the ISPRS Congress 2022 [Oral presentation].
2021
[8] A Data-driven Approach to Neural Architecture Search Initialization, René Traoré, Andrés Camero, Xiao Xiang Zhu, (under review, Journal) .
[7] Fitness Landscape Footprint: A Framework to Compare Neural Architecture Search Problems, René Traoré, Andrés Camero, Xiao Xiang Zhu, (under review, Journal).
[6] Lessons from Clustering of a Search Space: a Data-driven initialization technique to Search, René Traoré, Andrés Camero, Xiao Xiang Zhu, Workshop on 'Data Science meets Optimization', IJCAI 2021, [Oral presentation].
[5] Compact Neural Architecture Search for Local Climate Zones Classification, René Traoré, Andrés Camero, Xiao Xiang Zhu, ESANN 2021 , [Spotlight oral presentation].
2019
[4] DisCoRL: Continual Reinforcement Learning via Policy Distillation, René Traoré, Hugo Caselles-Dupré, Timothee Lesort, Te Sun, G. Cai, Natalia Díaz-Rodríguez, David Filliat, Workshop on 'Deep Reinforcement Learning', NeuRIPS 2019.
[3] Continual reinforcement learning deployed in real-life using policy distillation and sim2real transfer, René Traoré, Hugo Caselles-Dupré, Timothee Lesort, Te Sun, Natalia Diaz-Rodriguez, David Filliat, Workshop on 'Multi-Task and Lifelong Learning', ICML 2019.
[2] Decoupling feature extraction from policy learning: assessing benefits of state representation learning in goal-based robotics, Antonin Raffin, Ashley Hill, René Traoré, Timothee Lesort, Natalia Díaz-Rodríguez, David Filliat, Workshop on 'Structure & Priors in Reinforcement Learning' at ICLR 2019.
2018
[1] S-RL Toolbox: Environments, Datasets and Evaluation Metrics for State Representation Learning, Antonin Raffin, Ashley Hill, René Traoré, Timothee Lesort, Natalia Díaz-Rodríguez, David Filliat, Workshop on 'Deep Reinforcement Learning', NeurIPS 2018.
Scientific communication
2023:
[talk] International Conference on Optimization and Learning - Malaga, Spain
2022:
[poster] COnfiguration and SElection of ALgorithms (COSEAL) workshop - Virtual.
[poster] AutoML Conference 2022 - Virtual.
[talk + poster] ISPRS Congress 2022 - Nice, France.
[poster] Helmholtz Artificial Intelligence Conference 2022 - Dresden, Germany.
2021:
[poster] Machine Learning workshop (WAW) , German Aerospace Center - Virtual.
[poster] ESANN 2021 - Virtual.
[talk] Space and AI workshop, European Space Agengy & ECML-PKDD 2021 - Virtual.
[talk + poster] 'Data Science meets Optimization' workshop, IJCAI 2021 - Virtual.
[poster] Helmholtz Artificial Intelligence Conference 2021 - Virtual.
2019:
[poster] Eastern European Summer School 2019, University of Bucharest & DeepMind, Romania.
[poster] Data Science Summer School 2019, École Polytechnique, Palaiseau, France.
[poster] National Robotics Association (GDR Robotique), Télécom Paristech, Paris, France.
...
Color code for the events: International, National.
This selection is restricted to first author-only presentations, in venues where my office and I were not part of the organizing team.
Research software
State Representation Learning (SRL) for Goal-based Robotics [Youtube]
[Reinforcement Learning]
[Software 3 - RS3] Stable baselines : https://stable-baselines.readthedocs.io/en/master/
Mentoring
German Aerospace Center (DLR):
TUM Informatik - Master thesis intern - 2023
Technical University of Munich (TUM):
Munich Data Science Institute seminar - 2022 - [Five masters students / project] [proposal]
Remote Sensing seminar - 2019, 2020, 2021 - [One masters student / project].
ENSTA ParisTech:
Full-time mentoring of four interns (thesis) - 2019.
2 x Masters - 6 months each.
2 x Bachelors - 4 months each.
Affiliation of mentees: École Polytechnique, Shangai Jiaotong University, ENSTA ParisTech.