The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence) Spatiotemporal Data Analysis with Rose Yu - #508
Aug 9, 2021
In this engaging discussion, Rose Yu, an assistant professor at UC San Diego, delves into her groundbreaking work on machine learning for spatiotemporal data. She explains how integrating physical principles and symmetry enhances neural network architectures. The conversation covers innovative approaches in climate modeling, including turbulent prediction and the application of Physics Guided AI. Rose also addresses uncertainty quantification in models, crucial for applications like COVID-19 forecasting, showcasing the importance of confidence in predictions.
AI Snips
Chapters
Transcript
Episode notes
Spatiotemporal Data Analysis
- Rose Yu's research focuses on analyzing large-scale time-series and spatial-temporal data.
- She develops deep learning, tensor methods, and non-convex optimization techniques for this purpose.
Climate Modeling
- Climate models are computationally expensive, taking months to simulate.
- Yu's deep learning models emulate these simulations, speeding up climate, weather, and atmospheric predictions.
Physics-Guided AI
- Standard deep learning models struggle to capture the physical properties of phenomena like turbulence.
- Yu incorporates physical knowledge, like computational fluid dynamics and symmetries, to enhance predictions.