Felix Schur

I am a PhD student in Statistics at ETH Zurich under the supervision of Jonas Peters. My research interests lie in the intersection of Machine Learning and Statistics. I am particularly focused on Causality and Reinforcement Learning.

I completed my MSc at ETH Zurich, under the supervision of Andreas Krause and Peter Bühlmann. I wrote my master’s thesis on the topic of Lifelong Meta-Learning for Bayesian Optimization, which earned me the ETH Medal in recognition of my research.

I am also CTO at Askantis GmbH.

Contact: felix.schur [at] stat.math.ethz.ch

news

Mar 20, 2024	Teaching a Short Course on Causal Inference at HU Berlin together with Jonas Peters.
Mar 4, 2024	Presenting a poster on confounding in time series at the Machine Learning Summer School (MLSS) in Okinawa, Japan.
Oct 31, 2023	Reviewing for AISTAT.
May 8, 2023	LiBO presented at UAI 2023.

selected publications

2023

UAI

Lifelong Bandit Optimization: No Prior and No Regret

Felix Schur, Parnian Kassraie, Jonas Rothfuss, and 1 more author

39th Conference on Uncertainty in Artificial Intelligence (UAI), Aug 2023

Abs HTML PDF

Machine learning algorithms are often repeatedly applied to problems with similar structure over and over again. We focus on solving a sequence of bandit optimization tasks and develop LIBO, an algorithm which adapts to the environment by learning from past experience and becomes more sample-efficient in the process. We assume a kernelized structure where the kernel is unknown but shared across all tasks. LIBO sequentially meta-learns a kernel that approximates the true kernel and solves the incoming tasks with the latest kernel estimate. Our algorithm can be paired with any kernelized or linear bandit algorithm and guarantees oracle optimal performance, meaning that as more tasks are solved, the regret of LIBO on each task converges to the regret of the bandit algorithm with oracle knowledge of the true kernel. Naturally, if paired with a sublinear bandit algorithm, LIBO yields a sublinear lifelong regret. We also show that direct access to the data from each task is not necessary for attaining sublinear regret. We propose F-LIBO, which solves the lifelong problem in a federated manner.