CSE Community Seminar | October 4, 2024

Abstract

As the world works toward the UN Sustainable Development Goals, data gaps hinder our ability to measure progress and evaluate interventions. Meanwhile, rapid advances in computer vision and satellite imagery offer the opportunity to extract knowledge about the Earth at planetary scale. This talk will cover two applications of remote sensing and machine learning to generate large-scale insights. In the first, we use street-level imagery, satellite time series, and deep learning to map crop types for the first time across Thailand. In the second, we use deep learning to assess which US waters the Clean Water Act regulates and how Supreme Court and White House rulings change this regulation, providing a template for machine learning in policy interpretation.