Adrian Müller

Hi, I’m Adrian! I’m a second-year PhD student in Computer Science at ETH Zürich, fortunate to be advised by Niao He. As part of the ELLIS PhD program, I'm co-advised by Gergely Neu. My research interests are the foundations of reinforcement learning and theoretical machine learning more broadly.

I obtained my MSc in data science and my BSc in mathematics, both from ETH. Throughout my undergraduate studies, I had been very fortunate to be supported by the German Academic Scholarship Foundation („Studienstiftung“). I had also been an active long-term member of the ETH Student Sustainability Committee.

• 04/2026: I'm visiting IMSI at UChicago to participate in the Long Program on Theoretical Advances in Reinforcement Learning and Control.
• 07/2025: Presenting our paper Best of Both Worlds: Regret Minimization versus Minimax Play at ICML.
• 09/2024: I'm officially starting my PhD in the ODI lab!
• 07/2024: Together with Luca Viano, Pragnya Alatur and Anas Barakat, we organized the ETHZ-EPFL Summer School on Multi-Agent Reinforcement Learning.
• 07/2024: Presenting our paper Truly No-Regret Learning in Constrained MDPs as a Spotlight at ICML and at the Foundations of RL and Control Workshop.

Generative Modeling by Value-Driven Transport
Pablo Moreno-Muñoz⁺, Adrian Müller⁺, Gergely Neu⁺
Preprint
> arXiv link

Support Before Frequency in Discrete Diffusion
Adrian Müller, Antoine Gonon, Zebang Shen, Ya-Ping Hsieh, Niao He
Preprint
> arXiv link

Best of Both Worlds: Regret Minimization versus Minimax Play
Adrian Müller*, Jon Schneider*, Stratis Skoulakis*, Luca Viano*, Volkan Cevher
ICML, 2025
> arXiv link

Truly No-Regret Learning in Constrained MDPs
Adrian Müller, Pragya Alatur, Volkan Cevher, Giorgia Ramponi, Niao He
ICML, 2024 (Spotlight)
> arXiv link

• Spring 2025: Teaching Assistant for Theoretical Foundations of Reinforcement Learning
• 2024: Co-organizer of the ETHZ-EPFL Summer School on Multi-Agent Reinforcement Learning
• Fall 2020 & 2021: Teaching Assistant for Algorithms and Complexity