LANS Informal Seminar: Robert Gramacy

Seminar Title: Dynamic Trees for Response Surface Learning and Optimization
Speaker: Robert Gramacy, Booth Business School

Date/Time: 2011-03-23 15:00
Location: Bldg 240, 4301

Description:
This talk presents a new response surface methodology, dynamic trees, for learning, optimization, and sequential design — applications where Gaussian processes (GPs) have reigned supreme. Dynamic trees are thrifty on space and time: no need to store or invert big matrices. They are flexible: natively accommodating nonstationarity or heteroskedasticity. And they are inherently sequential, which means sequential design decisions (like for optimization) can be turned around quickly. Other benefits include the ability to deal with categorical predictors and responses, and to decompose partial dependencies by main, first order, and total effects. The talk will focus on the the dynamic tree methodology, inference by sequential Monte Carlo, and finally on design/optimization heuristics by active learning with illustrations. It will also highlight an R package implementing the methods, called dynaTree, which is available on CRAN. For more details see: http://arxiv.org/abs/0912.1586