Partnership between Climate AI Nordics and CLIMES

We are excited to share our partnership with CLIMES, the Swedish Centre for Impacts of Climate Extremes 🌍✨!
Read more on our /partners/!
We are excited to share our partnership with CLIMES, the Swedish Centre for Impacts of Climate Extremes 🌍✨!
Read more on our /partners/!
Published:
Chalmers University of Technology is hiring an Assistant or Associate Professor in Satellite Remote Sensing to join its leading Geoscience and Remote Sensing Division. The role focuses on advancing satellite-based Earth observation through research, teaching, and collaboration across academia and industry. Applications are open until 24 August 2025.
Event date:
Webinar with Markus Pettersson, Chalmers University of Technology. Machine learning models trained on Earth observation data, particularly satellite imagery, have recently shown impressive performance in predicting household-level wealth indices, potentially addressing chronic data scarcity in global development research. While these predictions exhibit strong predictive power, they inherently suffer from shrinkage toward the mean, resulting in attenuated estimates of causal treatment effects and thus limiting their utility in policy evaluations. Existing debiasing methods, such as Prediction-Powered Inference (PPI), require additional fresh ground-truth data at the downstream causal inference stage, severely restricting their applicability in data-poor environments. In this paper, we introduce and rigorously evaluate two novel correction methods—linear calibration correction and Tweedie's correction—that substantially reduce prediction bias without relying on newly collected labeled data. Our methods operate on out-of-sample predictions from pre-trained models, treating these models as black-box functions. Linear calibration corrects bias through a straightforward linear transformation derived from held-out calibration data, while Tweedie's correction leverages empirical Bayes principles to directly address shrinkage-induced biases by exploiting score functions derived from predicted outcomes. Through analytical exercises and experiments using Demographic and Health Survey (DHS) data, we demonstrate that both proposed methods outperform existing data-free approaches, can achieve significant reductions in attenuation bias and thus providing more accurate, actionable, and policy-relevant estimates. Our approach represents a generalizable, lightweight toolkit that enhances the reliability of causal inference when direct outcome measures are limited or unavailable.
Published:
Lund University is hiring an Assistant Professor in AI for Sustainable Transformation at the IIIEE, focusing on cutting-edge interdisciplinary research that bridges digital technologies and climate action. The position offers strong support for external funding, teaching, and collaboration with societal partners, with a pathway to permanent employment. Applications are open until 11 August 2025, and international researchers are especially encouraged to apply.
Published:
Amazing summary of the amazing member.