- This event has passed.
Machine learning for Oceanic & Atmospheric Sciences
January 7, 2022 @ 10:00 am - 3:00 pm
IDRE ECR Group is excited to announce Machine learning for oceanic & atmospheric sciences workshop with the following details:
Title: Machine Learning for Oceanic & Atmospheric Sciences
Date and Time: Friday, January 7, 2022 @10 AM (PST)
Abstract: Machine learning (ML) denotes a host of computational methods for inferring meaningful patterns in data. Advances over recent decades in ML methodologies, computational power, and dataset sizes have facilitated the rapid adaptation of these methods in practically every field of science and technology. ML allows learning directly from data to utilize hidden patterns in data where scientific theory is not fully developed or is too computationally intensive. ML has proved to be a highly useful tool in climate science, where traditional pattern recognition is confounded by high system complexity, multiscale interactions, chaos, and prohibitive datasets sizes.
The UCLA Machine Learning for Climate Workshop will host presentations from climate and ML scientists covering ML subjects of broad interest outside of climate as well as aspects that are especially of interest in climate science. Broad interest subjects will include discussions on choosing the right ML tools for the right task and tricks of the trade for common ML methodologies. Climate-related ML topics will consist of ML learning of spatio-temporal climate patterns, ML constrained by physical principles, and physical interpretability of learned patterns, as well as short research presentations.
Agenda: Friday, January 7, 2021
|Time (Pacific)||Presentation title and speaker|
|10:00-10:05||Welcome and Introduction|
|10:05-10:40||Inferring physics through self-supervised learning of climate data. Kaushik Srinivasan, UCLA|
|10:40-11:15||Explainable AI for Climate Science: Applications and Techniques. Kirsten Mayer, Colorado State University.|
|11:30-12:05||Tips for Successful Training of Deep Neural Networks. Bryor Snefjella, UCLA.|
|12:05-12:40||Machine learning for space weather prediction. Jacob Bortnik, UCLA.|
|13:40-14:10||Learning Stochastic Closures Using Ensemble Kalman Inversion. Jinlong Wu, Caltech.|
|14:10-14:45||Generative Modeling for Climate Science. Aditya Grover, UCLA.|
|14:45-15:00||Q&A with speakers|