My dissertation focused on three topics in probabilistic neural networks: Bayesian neural networks with fast approximations, incorporating nonparametric learning to normalizing flow models, and creating a flexible, edge-based method for generative graph neural networks.
Evan Ott and Sinead Williamson, 2023. Accepted to ICML workshop on Structured Probabilistic Inference & Generative Modeling.
Evan Ott and Sinead Williamson, 2022. Accepted at the I Can't Believe It's Not Better Workshop at NeurIPS 2022.
Over the past 30 years, an average of 85 people died each year in the US due to flash-floods, making them the most fatal severe weather condition. Particularly in Central Texas, the "most flash-flood prone area in the United States," we need to accurately predict rainfall. However, meteorologists continue to manually adjust state-of-the-art physical models based on experience, rather than objective methods.
This is where I come in.
My undergraduate thesis project uses neural networks and conditional random fields to better estimate rainfall in the Central Texas area. Rather than making future predictions, my project aims at determining the precise relationship between Doppler radar data and rainfall on the ground. For more details, be sure to check out the full write-up here and the code on GitHub. Special thanks goes to Dr. Michael Marder and Dr. Pradeep Ravikumar who supervised my research.