Adrienne Tuynman

I am a PhD student in the Inria Scool team at CRIStAL, where I work under the supervision of Emilie Kaufmann and Rémy Degenne. Before that, I was a student at the ENS Paris-Saclay and the MVA Master’s program.

I am particularly interested in identification problems in Multi-Armed Bandits and Reinforcement Learning. I like to craft algorithms that work with small sample-size for pure exploration problems, and that match the information lower bound for them.

Beyond that, I enjoy learning about statistical testing, sequential testing, safe statistics, and more broadly machine learning.

Publications

• Adrienne Tuynman, Rémy Degenne (2025). The Batch Complexity of Bandit Pure Exploration. ICML 2025.
• Adrienne Tuynman, Rémy Degenne, Emilie Kaufmann (2024). Finding good policies in average-reward Markov Decision Processes without prior knowledge. NeurIPS 2024.

Pre-publication

• Victor Boone, Adrienne Tuynman (2025). Towards Blackwell Optimality: Bellman Optimality Is All You Can Get
• Adrienne Tuynman, Ronald Ortner (2022). Transfer in Reinforcement Learning via Regret Bounds for Learning Agents

News

• [September 2025] I was in EWRL2025 in Tübingen to present my batched bandits paper.
• [September 2025] With my coworker Timothée Mathieu, we prepared an entry for the Summer of Math Exposition competition. Check out our animated video about how to conduct safe tests.
• [July 2025] I will be back in Vancouver to present our paper about batched bandits at ICML 2025.
• [June 2025] I was at COLT.
• [January 2025] I presented my work at the virtual RL seminar. You can watch the recording on YouTube.
• [December 2024] My paper about average-reward RL has been accepted at NeurIPS 2024 in Vancouver. See you there!
• [October 2024] I will be at the EWRL2024 in Toulouse to present my work on average-reward RL!
• [26-27 October 2023] I participated to the RJMI at the Inria center of Lille, introducing a group of highschool girls to the world of scientific research, and supervising them as they worked through an open mathematical problem.
• [October 2023] I started my PhD about fixed budget best arm identification with Emilie Kaufmann and Rémy Degenne at the Inria center of Lille in the Scool team.
• [Spring and summer 2023] I conducted my final Master’s internship with Emilie Kaufmann and Rémy Degenne at the Inria center of Lille in the Scool team about best policy identification in average reward RL.
• [September 2022] We presented our paper about transfer learning at EWRL 2022.
• [Spring and summer 2021] I conducted an internship with Ronald Ortner at the Montanuniversität Leoben about transfer learning in average reward RL.
• [Summer 2020] I conducted a month long internship with Philippe Preux at the Inria center of Lille in the SequeL team.

Teaching

Teaching material

Check out me and Timothée Mathieu’s animated video about performing safe tests here.

2023-2026

I am the TA for the “Bases mathématiques” course in the first semester again, and for the Cryptography course in the second semester for the mathematics Master’s of the Lille University.