Till Freihaut

Bio

Hi! I am a PhD student interested in the mathematical aspects of Reinforcement Learning, particularly in multi-agent systems with a focus on imitation learning. I am fortunate to be advised by Giorgia Ramponi at the University of Zurich. Additionally, I am an affiliated doctoral student at the ETH AI Center.

Previously, I completed my master's degree in mathematics at the University of Mannheim (with honors), where I was co-advised by Claire Vernade , Leif Döring and Simon Weissmann. In my thesis I worked on (contextual) bandits and importance sampling.

News

I am happy to announce that I received the UZH Candoc Grant.
May 2026: I started an internship as a machine learning researcher at Point72.
Multi-agent imitation learning with function approximation: Linear Markov games and beyond got accepted to ICML!
Rate optimal learning of equilibria from data got accepted to AISTATS!
New Preprint with Luca, Emanuele, Volkan, Matthieu and Giorgia: Rate optimal learning equilibria from data
2/2 papers accepted at NeurIPS2025, including one spotlight!
Provably Efficient Multi-Agent Imitation Learning: Learning equilibria from data has received a spotlight at the RLC workshop RLBrew!
Three papers accepted at EWRL18.
We are organizing a NeurIPS workshop!
New Preprint with Luca, Volkan, Matthieu and Giorgia: Provably Efficient Multi-Agent Imitation Learning: Learning equilibria from data
New Preprint resulting from my master thesis with Simon, Claire, Giorgia and Leif
First PhD project: On Feasible Rewards in Multi-Agent Inverse Reinforcement Learning accepted at RLDM 2025
May 2024: I started my PhD at the University of Zurich.

Publications

Most recent publications on Google Scholar.
^‡ indicates equal contribution.

All

Multi-agent imitation learning with function approximation: Linear Markov games and beyond

Luca Viano^‡, Till Freihaut^‡, Emanuele Nevali,Volkan Cevher, Matthieu Geist, Giorgia Ramponi

ICML 2026.

Paper

Rate optimal learning equilibria from data

Till Freihaut^‡, Luca Viano^‡, Emanuele Nevali,Volkan Cevher, Matthieu Geist, Giorgia Ramponi

AISTATS 2026.

Paper

Learning Equilibria from Data: Provably Efficient Multi-Agent Imitation Learning

Till Freihaut^‡, Luca Viano^‡, Volkan Cevher, Matthieu Geist, Giorgia Ramponi

NeurIPS 2025.

Paper

On Feasible Rewards in Multi-Agent Inverse Reinforcement Learning

Till Freihaut, Giorgia Ramponi

NeurIPS 2025.

Paper

Barycenter Policy Design for Multiple Policy Evaluation

Simon Weissmann^‡, Till Freihaut^‡, Claire Vernade, Giorgia Ramponi, Leif Döring

EWRL 2025.

Paper

Multi-agent imitation learning with function approximation: Linear Markov games and beyond

Luca Viano^‡, Till Freihaut^‡, Emanuele Nevali,Volkan Cevher, Matthieu Geist, Giorgia Ramponi

ICML 2026.

Paper

Rate optimal learning equilibria from data

Till Freihaut^‡, Luca Viano^‡, Emanuele Nevali,Volkan Cevher, Matthieu Geist, Giorgia Ramponi

AISTATS 2026.

Paper

Learning Equilibria from Data: Provably Efficient Multi-Agent Imitation Learning

Till Freihaut^‡, Luca Viano^‡, Volkan Cevher, Matthieu Geist, Giorgia Ramponi

NeurIPS 2025.

Paper

On Feasible Rewards in Multi-Agent Inverse Reinforcement Learning

Till Freihaut, Giorgia Ramponi

NeurIPS 2025.

Paper

Barycenter Policy Design for Multiple Policy Evaluation

Simon Weissmann^‡, Till Freihaut^‡, Claire Vernade, Giorgia Ramponi, Leif Döring

EWRL 2025.

Paper

Vitæ

Full Resume in PDF.

Website Design

Thank you Martin Saveski for this awesome template.