Parameter Estimation and Interpretability in Bayesian Mixture Models
Offered By: VinAI via YouTube
Course Description
Overview
Explore the intricacies of parameter estimation and interpretability in Bayesian mixture models through this comprehensive seminar series. Delve into the research of Long Nguyen, an associate professor at the University of Michigan, as he examines posterior contraction behaviors for parameters in Bayesian mixture modeling. Investigate two types of prior specification: one with an explicit prior distribution on the number of mixture components, and another placing a nonparametric prior on the space of mixing distributions. Learn how these approaches yield optimal rates of posterior contraction and consistently recover unknown numbers of mixture components. Analyze the impact of model misspecification on posterior contraction rates, with a focus on the crucial role of kernel density function choices. Gain insights into the tradeoffs between model expressiveness and interpretability in mixture modeling, equipping yourself with valuable knowledge for statistical modeling in various applications.
Syllabus
Seminar Series: Parameter Estimation & Interpretability in Bayesian Mixture Models
Taught by
VinAI
Related Courses
Introduction to Probability, Statistics, and Random ProcessesUniversity of Massachusetts Amherst via Independent Bayesian Statistics
Duke University via Coursera Bayesian Statistics: From Concept to Data Analysis
University of California, Santa Cruz via Coursera Improving your statistical inferences
Eindhoven University of Technology via Coursera Bayesian Statistics: Techniques and Models
University of California, Santa Cruz via Coursera