2x PhD Positions as part of the SNSF Project -From Alps to Arctic: Satellite-based Assessment of Forest Canopy Height across Decades- 80 %

About Us

The University of Zurich, Switzerland's largest university, is home to a diverse array of attractive roles across multiple subject areas and professional fields. With approximately 10,000 employees and an impressive array of professional apprenticeship streams, the University cultivates an inspiring working environment centered on cutting-edge research and top-tier education. Join us and leverage your talent and skills in this dynamic setting.

Your Responsibilities

As part of the SNSF project, the EcoVision Lab will emphasize advanced forest parameter estimation, focusing particularly on canopy height at a granular level. The work involves developing innovative deep learning and computer vision techniques to convert extensive remote sensing imagery from various satellite missions into detailed maps of canopy height and other forest parameters, tracking their evolution over time.

Your research will encompass:

• Developing deep learning models for satellite image time-series analysis and domain adaptation.
• Creating models for (guided) super-resolution of historical satellite imagery.
• Producing calibrated uncertainty estimates for all model outputs.
• Training models on heterogeneous data sources (including Landsat, Sentinel-2, SPOT, Corona) and exploring multimodal combinations of these data sources.

Research Freedom & Methodological Innovation

This project provides substantial freedom to explore influential methodological directions in modern AI, incorporating self-supervised learning, multimodal learning, guided super-resolution, uncertainty estimation, and time-series regression. The goal is to achieve high-impact publications in prominent machine learning venues as well as leading interdisciplinary journals.

Why Join?

These positions are an opportunity to become part of the EcoVision Lab, a vibrant and stimulating place for engaging in research on deep learning applied to ecological contexts. Benefits include:

• Collaboration with leading research teams in machine learning, computer vision, data science, remote sensing, and historical remote sensing image interpretation.
• The chance to merge groundbreaking AI research with real-world environmental implications focused on a largely under-explored research topic.
• Access to diverse, large-scale historical satellite image archives.

Your Profile

We seek highly motivated individuals passionate about advancing machine learning while driving significant environmental impact. Candidates should be curious, rigorous, and enthusiastic about developing innovative ideas and high-quality research software while tackling complex problems and collaborating across disciplines.

An ideal candidate will possess:

• An excellent Master's degree (M.Sc. or equivalent) in Computer Science, Machine Learning, Data Science, or a related field (e.g., Electrical Engineering, Applied Mathematics).
• A solid understanding of mathematics and machine learning.
• Extensive programming experience, preferably in Python.
• Strong background in deep learning and computer vision.
• A keen interest in applying advanced machine learning techniques to ecological and geospatial data.
• Fluency in English (both written and spoken).

Familiarity with topics such as self-supervised learning, domain adaptation, transfer learning, multimodal learning, and uncertainty estimation is advantageous but not a strict requirement. We are dedicated to fostering a diverse and inclusive research environment and encourage candidates from all backgrounds to apply, especially those who may not meet every listed criterion but demonstrate strong motivation and potential.