PhD position in Explainable AI for Rare Crop Phenotype Discovery

This PhD project sits at the intersection of artificial intelligence, functional ecology, and agroecology, aiming to address the challenge of identifying rare crop phenotypes from scattered scientific data. The candidate will work on building a global database of functional traits, developing anomaly detection methods in high-dimensional spaces, and creating explainable AI frameworks to distinguish biological discoveries from data errors. The research is hosted in a highly interdisciplinary environment in Montpellier, involving three specialized research groups. The successful applicant will contribute to both AI methodology and the understanding of crop functional diversity, with potential applications across various scientific domains.

• Develop and maintain a global database of functional traits for cultivated plants.
• Apply Large Language Models and text mining techniques to extract information from scientific literature.
• Design and implement machine learning algorithms for anomaly and rare-event detection in high-dimensional trait spaces.
• Create computational argumentation frameworks to evaluate the credibility of candidate rare phenotypes.
• Collaborate with interdisciplinary teams across plant ecophysiology and AI research groups.
• Document research findings and contribute to academic publications and conference presentations.