Spatiotemporal Data Analysis with Rose Yu - #508
Published:Aug 9, 2021 18:08
•1 min read
•Practical AI
Analysis
This article summarizes a podcast episode featuring Rose Yu, an assistant professor at UC San Diego. The focus is on her research in machine learning for analyzing large-scale time-series and spatiotemporal data. The discussion covers her methods for incorporating physical knowledge, partial differential equations, and exploiting symmetries in her models. The article highlights her novel neural network designs, including non-traditional convolution operators and architectures for general symmetry. It also mentions her work on deep spatio-temporal models. The episode likely provides valuable insights into the application of machine learning in climate, transportation, and other physical sciences.
Key Takeaways
- •Rose Yu's research focuses on machine learning for spatiotemporal data analysis.
- •She incorporates physical knowledge and partial differential equations in her models.
- •Her work includes novel neural network designs with non-traditional convolution operators.
Reference
“Rose’s research focuses on advancing machine learning algorithms and methods for analyzing large-scale time-series and spatial-temporal data, then applying those developments to climate, transportation, and other physical sciences.”