Abstract: In this workshop, we'll explore essential PyTorch tasks such as creating custom models and processing data, emphasizing how to integrate models with existing frameworks to boost efficiency. Participants will then use these skills to build, train, and estimate uncertainty in probabilistic models. This process will culminate in the application of these models within a Bayesian optimization framework, illustrating a complete workflow.
Speaker Biography: Felix is a fourth-year PhD student in Statistics at the University of Wisconsin–Madison, his research focus is on nearest neighbor based approximations for Gaussian processes, uncertainty quantification in deep learning, and Bayesian optimization. He applies these methodologies to various scientific applications. Before graduate school, Felix worked for two years in the Statistical Engineering Division at NIST. He started his PhD at Texas A&M, then moved to the University of Wisconsin–Madison with his advisor. In summer 2023, he interned at the Toyota Research Institute in Los Altos, CA, developing automated methods for processing battery cell diagnostic data.
