Presentation

This is the website of Inria team-project named Scool.

Scool was created on November 1st, 2020. Scool is the follow-up of SequeL.

In a nutshell, the research topic of Scool is the study of the sequential decision making problem under uncertainty. Most of our activities are related to either bandit problems, or reinforcement learning problems. Through collaborations, we are working on their application in various fields, mainly: health, agriculture and ecology, sustainable development. See our Projects page for more information.

Scool is constantly looking for people to hire: post-docs, engineers, permanent researchers, Ph.D. students. See our Job offers for more information.

Recent news:

  • We organize the Worskhop on Interpretable Policies in Reinforcement Learning (InterpPol) at the Reinforcement Learning Conference, Amherst, Aug. 2024.
  • Émilie is awarded the bronze medal from CNRS. This medal is an important form of recognition for the men and women who make science progress.
  • Debabrota gives a tutorial at IJCAI 2023 on “Auditing Bias of Machine Learning Algorithms: Tools and Overview
  • 1 ANR JCJC project has been accepted in 2023 and will start in early 2024:
    • NeuRL (Neuro-Incremental Reinforcement Learning from Human Preferences) headed by Riad Akrour.
  • Riad’s Inria exploratory action AuDaCiTi (Autonomous Data Collection and Labelling Through Interaction) has been selected in May 2023.
  • 3 ANR projects have been accepted and start in early 2023, ending in 2026:
    • FATE (Frugal and Adaptive Tests) headed by Rémy Degenne,
    • REPUBLIC (Towards responsible AI with reinforcement learning under constraints) headed by Debabrota Basu,
    • BIP-UP (Bandits Improve Patient follow-UP) headed by Philippe Preux, in collaboration with Prof. F. Pattou, INSERM and Hospital of Lille.

    We will hire 3 Ph.D. students, post-docs and engineers.

  • Rémy co-organized the Complex feedback in online learning workshop at ICML, held in Baltimore in July 2022.
  • Our weight prediction tool dedicated to patients who had bariatric surgery is on-line. This results from our collaboration with F. Pattou’s team at INSERM and Hospital of Lille.
  • We release gym-DSSAT, a gym environment to play the crop management game is available here accompanied by a report. gym-DSSAT is based on DSSAT, a high fidelity crop simulator.

More news on our News page.


A picture of some Scool members in March 2023, just outside the Inria building, Lille, France.

Comments are closed.