PhD on Conditional Generative Modelling of Local High-Impact Events under Structured Scenarios

The University of Bern invites applications for a PhD project titled "Conditional Generative Modelling of Local High‑Impact Events under Structured Scenarios". The position is based in the Institute of Mathematical Statistics and Actuarial Science (IMSV) and belongs to the NCCR CLIM⁺ programme, which is funded by the Swiss National Science Foundation and focuses on climate extremes and societal adaptation. As a doctoral researcher you will join the Uncertainty Quantification and Spatial Statistics group led by Prof. David Ginsbourger. Your daily work will support the work package "Local Risks: Impacts and Adaptation", involving tasks such as down‑scaling and debiasing input data for hydrological models. You will regularly interact with collaborators from ETH Zurich, WSL, MeteoSwiss, EPFL Valais‑Wallis and other partners, contributing to interdisciplinary studies and engaging with external stakeholders. The role requires a solid academic foundation and technical expertise. Candidates must hold an MSc in statistics or a closely related discipline and possess a strong mathematical background. Proficiency in programming, preferably Python and/or R, is essential, as is knowledge of generative machine‑learning methods. Experience handling large‑scale hydrological, meteorological or climate data sets is expected, together with very good oral and written English communication skills. A genuine motivation to work in an international, interdisciplinary team and a collaborative mindset are also required. The position offers a vibrant research environment within an open‑minded institute where statistical theory meets real‑world data challenges. NCCR CLIM⁺ promotes equal opportunity and values diversity of thought and background. You will be part of a network that includes leading universities and research institutes across Switzerland, contributing to climate‑resilient solutions. What the role asks for: - MSc in statistics or related field - Strong mathematical background - Proficient programming in Python or R - Knowledge of generative machine learning - Experience with large hydrological or climate datasets - Excellent oral and written English skills