Maurice Weber

Maurice Weber

PhD candidate in Computer Science

ETH Zürich

About me

I am a fourth year PhD student at ETH Zurich where I am co-supervised by Ce Zhang and Martin Vechev.

My research focuses on developing robustness guarantees for different aspects of machine learning pipelines, including robustness against data poisoning attacks, semantic transformations and distributional shifts. Next to my theoretical work I am extremely excited about the interplay between data quality and large language models. I am also a firm believer in open-source AI, where I am fortunate to have been able to contribute to the RedPajama-1T dataset.

I hold a BSc and an MSc in Mathematics, both from ETH Zürich, where I focused on statistics and machine learning and graduated in 2019 with distinction.

Feel free to contact me by email, on Twitter or on LinkedIn.

Publications & Preprints

RAB: Provable Robustness against Backdoor Attacks
Certifying Some Distributional Fairness with Subpopulation Decomposition
Predicting Properties of Quantum Systems with Conditional Generative Models
Toward reliability in the nisq era: Robust interval guarantee for quantum measurements on approximate states
Certifying Out-of-Domain Generalization for Blackbox Functions
The AI Neuropsychologist: Automatic scoring of memory deficits with deep learning
TSS: Transformation-Specific Smoothing for Robustness Certification
Optimal provable robustness of quantum classification via quantum hypothesis testing
Ease. ML: A Lifecycle Management System for MLDev and MLOps
Observer Dependent Lossy Image Compression
Towards device-agnostic mobile cough detection with convolutional neural networks

Students

I am fortunate to have adivsed the excellent MSc students

  • Anton Alexandrov (2023, ETH Zurich)
  • Valdemar Thanner (2023, ETH Zurich)
  • Georgios Tsolakis (2023, ETH Zurich)
  • Felix Asanger (2021, ETH Zurich, now @ Zurich Insurance)
  • Himankar Sharma (2021, ETH Zurich, now @ Scandit)

Contact