Stefan Sperlich

Prof.

University of Geneva

Switzerland

Research Interests

Explore related searches

Professors in Switzerland Professors in Statistics Statistics professors in Switzerland

Contact this professor

ORCID

Google Scholar

Positions (1)

Slava Voloshynovskiy

University Name

University of Geneva

3 weeks ago

Fully Funded PhD Positions in Foundation Models for Scientific Discovery and AI Applications

University of Geneva (Switzerland) is recruiting 9 fully funded PhD positions in the FUNDIS (Foundation Models for Scientific Discovery) project. The positions focus on foundation models , self-supervised learning , world models , explainable AI , bio-inspired AI , and AI for scientific discovery , with applications across astronomy , particle physics , climate science , digital humanities , and global governance . Students will work in an outstanding multidisciplinary research environment with close collaboration across AI and domain sciences, strong publication opportunities, and international scientific collaborations. The post highlights competitive Swiss PhD salaries and co-supervision by an interdisciplinary team of PIs including Slava Voloshynovskiy, François Fleuret, Roland Bouffanais, Daniel Schaerer, Pascal Oesch, Stéphane Paltani, Tobias Golling, Sebastian Engelke, Stefan Sperlich, Béatrice Joyeux-Prunel, and Stas Smirnov. This is a PhD-level opening only. No specific eligibility details such as degree background, GPA, or language requirements are stated in the post. Application deadline: 30 August 2026 .

3 weeks ago

Articles (1)

A Note on Simultaneous Confidence Intervals for Direct, Indirect and Synthetic Estimators

Direct, indirect and synthetic estimators have a long history in official statistics. While model-based or model-assisted approaches have become very popular, direct and indirect estimators remain the predominant standard and are therefore important tools in practice. This is mainly due to their simplicity, including low data requirements, assumptions and straightforward inference. With the increasing use of domain estimates in policy, the demands on these tools have also increased. Today, they are frequently used for comparative statistics. This requires appropriate tools for simultaneous inference. We study devices for constructing simultaneous confidence intervals and show that simple tools like the Bonferroni correction can easily fail. In contrast, uniform inference based on max-type statistics in combination with bootstrap methods, appropriate for finite populations, work reasonably well. We illustrate our methods with frequently applied estimators of totals and means.

Article Link

Year:

2024