Prof. Dr. Beate Sick

Group Leader

In 2026, Prof. Dr. Beate Sick joined the Thurgauer An-Institute for Digital Transformation (TIDIT) in Kreuzlingen, where she leads the group for Probabilistic AI. Her interdisciplinary research bridges statistics, artificial intelligence, and causality, with a particular focus on multi-modal and transparent models in applied biomedical research.

Before joining TIDIT, she was heading the group of bioinformatics in the DNA Array Facility at University of Lausanne and afterwards  spending more than 10 years as a Professor at the Zurich University of Applied Sciences (ZHAW), where she held the Chair of Applied Statistics and led a team of students and young researchers in close collaboration with collaboration partners from different Universities, hospitals and industry. Since 2014, she is also affiliated with the University of Zurich, teaching in the Master’s program in Biostatistics and Epidemiology and conducting collaborative research at the interface of AI and biostatistics in close partnership with medical researchers. In addition, she lectures at ETH Zurich on biostatistics, multivariate statistics, and deep learning.

Competencies

  • Probabilistic AI
  • Causal Modeling
  • Biostatistics
  • Applied Research
Sick-Beate
Contact
Prof. Dr. Beate Sick
Group Leader
Probabilistic AI Research Group
Sick-Beate
Contact
Prof. Dr. Beate Sick
Group Leader
Probabilistic AI Research Group

Probabilistic AI Research Group

Despite the groundbreaking progress in deep learning and AI, these technologies often struggle to quantify uncertainties and lack interpretability. This limitation is evident in issues like hallucinations in large language models, where generated outputs can be plausible yet incorrect. Such challenges are critical in various contexts, including load forecasting and medical diagnostics.
Probabilistic AI

Probabilistic AI Research Group

Despite the groundbreaking progress in deep learning and AI, these technologies often struggle to quantify uncertainties and lack interpretability. This limitation is evident in issues like hallucinations in large language models, where generated outputs can be plausible yet incorrect. Such challenges are critical in various contexts, including load forecasting and medical diagnostics.
Probabilistic AI

Publications

Loading publications...

News

February 19, 2026

Eröffnung des I+D Campus 2026: mit dem TIDIT als Teil des Campus

January 6, 2026

Agentic AI praktisch erleben – Jetzt anmelden!