Early Career Researcher (ECR) of the Month (Feb 2024) – Lily-belle Sweet

By Lou Brett, University of Strathclyde

? We want to hear from you and your exciting research! Always wanted to make your research more visible? If you’re interested in joining the CompoundNET early career network, or even featuring as an EC of the month, this is your chance to ?️ fill out this form!

Our second Early Career Researcher (ECR) of the Month for February 2024, Lily-belle Sweet!

Lily-belle is currently completing her PhD at the Helmholtz Centre for Environmental Research -UFZ Leipzig. Lily-belle’s research topic explores using interpretable or explainable machine learning to identify compounding climate drivers of agricultural yield failure. 

Lily-belle is currently using data-driven methods to identify climate drivers of impacts such as agricultural yield shocks, with particular focus on the interactions between driving events. Lily-belle is also one of the coordinators of the AgMIP Machine Learning team (AgML), which was initiated during a Short-Term Scientific Mission to NASA GISS in New York (Check out her blog post here!). Her most recent work explores the importance of cross-validation strategy when using machine learning on spatiotemporal data, for model performance and interpretation via permutation feature importances (a frequently-used tool for identifying drivers of climate impacts).

If you are interested in reading this paper further, please find the link beneath:

Cross-validation strategy impacts the performance and interpretation of machine learning models, Sweet et al. (2023), Artificial Intelligence for the Earth Systems, URL: https://doi.org/10.1175/AIES-D-23-0026.1 

Post edited by Pauline Rivoire. Photo credits: Pauline Rivoire (top image), Lily-belle Sweet (portrait).