(EN) Open-Source Earth observation datasets for deep structured learning-based monitoring
Monitoring oceans, agricultural lands, forests, and cities using satellite imagery is a key area of ongoing research. Particularly, the rise of open-source large-scale Earth observation datasets for this purpose facilitates the development of machine learning models like artificial neural networks for assessment of landscapes. In this talk, we discuss how we can train deep learning models and especially harness multitemporal data and change detection to detect damage after natural disasters, inform sustainable agriculture, monitor wildlife, and more.
Thomas ChenU.S. Technology Policy Committee
Thomas Chen is a machine learning researcher who serves on the U.S. Technology Policy Committee of the Association for Computing Machinery. Previously, Thomas has presented work at a number of conferences, workshops, and meetings, from NeurIPS workshops, to Applied Machine Learning Days, to the Open Data Science Conference, to Machine Learning Week Europe.