
Institute · Bordeaux, France
FlowersInria and Univ. Bordeaux
Flowers is an Inria team in Bordeaux studying curiosity-driven learning in humans and machines, led by Pierre-Yves Oudeyer.
Bordeaux, France
- Type
- Institute
- Based in
- Bordeaux, France
- Featured projects
- 4
About
The Flowers team investigates the mechanisms of sensorimotor, cognitive and social development in humans and machines. It is housed at Inria Bordeaux-Sud-Ouest and Univ. Bordeaux and is headed by Pierre-Yves Oudeyer, a research director known for work on artificial curiosity and intrinsic motivation.
The team's GitHub organization, opened in 2012, publishes research code such as the Explauto exploration library and testbeds for automatic curriculum learning in deep reinforcement learning. More recent releases ground language models in interactive environments using online reinforcement learning and provide the Lamorel library for that purpose.
Mission and ambitionsTo understand how open-ended, curiosity-driven learning works and to build artificial agents that develop and learn in similar ways.
Known for
- 01
Curiosity-driven and developmental AI
- 02
Pierre-Yves Oudeyer's research team
- 03
Grounding LLMs with reinforcement learning
Milestones
- 2012
GitHub organization opened
The team began publishing research code under flowersteam.
- 2014
Explauto library
An autonomous exploration library for developmental robotics was released.
- 2019
teachDeepRL
Code for teacher algorithms in deep reinforcement learning was published.
- 2021
TeachMyAgent
A testbed platform for automatic curriculum learning methods in deep RL was released.
- 2023
Grounding LLMs with online RL and Lamorel
Code for functionally grounding LLM knowledge in BabyAI-Text and a library for RL with LLMs were published.
Open source
We perform functional grounding of LLMs' knowledge in BabyAI-Text (277 stars)
Lamorel is a Python library designed for RL practitioners eager to use Large Language Models (LLMs). (247 stars)
Open-source repository. (94 stars)
TeachMyAgent is a testbed platform for Automatic Curriculum Learning methods in Deep RL. (78 stars)
Notable releases
Openness
Research code is released openly in the flowersteam GitHub organisation, typically alongside papers.
People
- Pierre-Yves OudeyerTeam head, Inria research director
Listed from the lab's own pages and public profiles.
Contact and community
Contact through the team website.
Sources
Compiled from the lab's public pages, Hugging Face and GitHub. Last checked 2026-11-17. Something wrong or out of date? Tell us on Discord.



