Alexandre Heuillet

PhD student in Deep Learning at IBISC Lab, Université Paris-Saclay. Incoming Visiting Student at MIT.

Massachusetts Institute of Technology

IBISC Laboratory

Université Paris-Saclay



  • Deep Learning
  • NAS, AutoML
  • GANs
  • Computer Vision
  • Reinforcement Learning


  • PhD in Computer Science, Deep Learning, 2023

    Univsersité Paris-Saclay

  • MSc in Software Engineering with an Artificial Intelligence speciality, 2020

    École Nationale Supérieure d'Électronique, Informatique, Radiocommunications et Mathématiques-Mécanique de Bordeaux (ENSEIRB-MATMECA)

  • Two-years intense Mathematics and Physics curriculum (CPGE), 2017

    Cycle Préparatoire de Bordeaux (CPBx)



Teaching Assistant

Université Paris Saclay

Oct 2020 – Oct 2023 Paris, France
I taught Deep Learning and Algorithmics classes in English for international M1 and M2 students (ITEE, MMVAI). I also taught C programming for L2 students.

Research Intern

Groupe Renault

Feb 2020 – Jul 2020 Paris, France
I worked on Deep Learning (GANs, mostly Few-Shot Vid2Vid from Nvidia Research) applied to Advanced Driver Assistance Systems (ADAS). My project aimed to synthetize photorealistic street videos from semantic masks (segmented images) and environmental descriptions (e.g. weather or luminance conditions). This helped test onboard cameras mounted on autonomous vehicles.

Research Intern

Kurazume Laboratory, Graduate School of Electrical Engineering and Computer Science, Kyushu University

Jun 2019 – Aug 2019 Fukuoka, Japan
Worked on a Mixed Reality (Microsoft Hololens) app to reproduce human motions on holograms. This was part of a project that aimed to provide a better way to teach the Humanitude care method for elderly people.


Omnitech Security

Jun 2018 – Jul 2019 Bordeaux, France
Intern in a security software company. Unitary and integration tests conduction and redaction of PHP scripts to remote control domotic equipment.

Head of Information Systems

Aquitaine Électronique Informatique

Apr 2018 – Apr 2019 Bordeaux, France

Junior-Enterprise IT manager. Responsibilities included:

  • Maintenance and development of various software and websites (Worpress, CRM, Java clients…)
  • IT consulting
  • Team and project management


Explainability in Reinforcement Learning

I participated in the redaction of a survey on state-of-the-art explainability methods for reinforcement learning. We also implemented a novel technique for explaining collaborative tasks in multiagent environments. Check out our code here!


As part of a school project, I created a Web app (using Django) for INRIA researchers that helped them train their RNNs (Recurrent Neural Networks) and visualize results in an intuitive way. Check out the code here!

Othello Player

I implemented an AlphaZero-like Othello player as a school project. It is based on a Monte-Carlo Tree Search algorithm improved by a neural network powered move prediction.

Recent Publications

  • A. Heuillet, H. Tabia, H. Arioui, and K. Youcef-Toumi. D-DARTS: Distributed Differentiable Architecture Search. arXiv, 2022.
  • A. Heuillet, F. Couthouis and N. Diaz Rodriguez. Collective eXplainable AI: Explaining Cooperative Strategies and Agent Contribution in Multiagent Reinforcement Learning With Shapley Values. IEEE Computational Intelligence Magazine, 2022.
  • A. Heuillet, F. Couthouis and N. Diaz Rodriguez. Explainability in Reinforcement Learning. Knowledge-Based Systems, 2021.