Representation Learning and Custom Loss Functions for Atmospheric Data - Christian Lessig
Kavli Institute for Theoretical Physics via YouTube
Overview
Explore representation learning and custom loss functions for atmospheric data in this conference talk from the Machine Learning for Climate KITP conference. Delve into how big data and machine learning algorithms are revolutionizing climate science, enabling detailed analysis of complex Earth system processes. Discover how these advanced techniques can lead to improved understanding of multi-scale interactions in the physical, chemical, and biological realms. Learn about the potential for descriptive inference to drive new theories and validate existing ones in climate research. Gain insights into how data-driven approaches, when combined with modeling experiments and robust parameterizations, can address challenging questions in climate science. Understand the importance of interdisciplinary collaboration in advancing climate change research and the role of this conference in setting the stage for future progress in the field.
Syllabus
Representation learning and custom loss functions for atmospheric data â–¸ Christian Lessig
Taught by
Kavli Institute for Theoretical Physics