Location: 232A Withers Hall, NC State Main Campus

Title: Building faster and more expressive BART models

Presenter: Sameer Deshpande

Abstract:

Bayesian Additive Regression Trees (BART) is a highly effective nonparametric regression model that approximates unknown functions with a sum of binary regression trees. Most implementations of BART are based on trees that (i) recursively partition continuous inputs one variable at a time; (ii) one-hot encode categorical predictors; and (iii) represent piecewise constant functions. These implementations are fundamentally limited in their ability to learn complex decision boundaries that are not aligned with coordinate axes; to “borrow strength” across multiple groups; to leverage structural relationships between multiple categorical predictors (e.g., adjacency and nesting); and to estimate smooth functions.