The article by Picheny et al.(2013) is very interesting, and I believe that the methodologies developed here are innovative and should be useful for computer experiment practitioners. I congratulate the authors for writing such a nice article. Although I like most of the ideas in this article (especially the time allocation strategy), the formulation of the improvement function may need more justification, and this is the main focus of this discussion. To recall, the key assumptions of Picheny et al. are as follows:

Pritam Ranjan

Operation Management and Quantitative Techniques

IIM Indore seeks to be a contextually-relevant business school with world-class academic standards that develops socially-conscious managers, leaders and entrepreneurs. IIM Indore is committed to: Excellence in management education, research, and training, Use of contemporary participant-centric pedagogies and teaching methods, A presence in emerging segments of management education.

IIM Indore

Comment: EI Criteria for Noisy Computer Simulators

Technometrics

In computer experiments, a mathematical model implemented on a computer is used to represent complex physical phenomena. These models, known as computer simulators, enable experimental study of a virtual representation of the complex phenomena. Simulators can be thought of as complex functions that take many inputs and provide an output. Often these simulators are themselves expensive to compute, and may be approximated by "surrogate models" such as statistical regression models. In this paper we consider a new kind of surrogate model, a Bayesian ensemble of trees (Chipman, George, &amp; McCulloch, 2010), with the specific goal of learning enough about the simulator that a particular feature of the simulator can be estimated. We focus on identifying the simulator's global minimum. Utilizing the Bayesian version of the expected improvement criterion (Jones, Schonlau, &amp; Welch, 1998), we show that this ensemble is particularly effective when the simulator is ill-behaved, exhibiting nonstationarity or abrupt changes in the response. A number of illustrations of the approach are given, including a tidal power application. extcopyright 2012 Statistical Society of Canada.

Fulltext

Canadian Journal of Statistics

Sequential design for computer experiments with a flexible Bayesian additive model

Quantile-Based Optimization of Noisy Computer Experiments With Tunable Precision

Minimum-Aberration Two-Level Split-Plot Designs

Institute of Mathematical Statistics Lecture Notes - Monograph Series New developments and applications in experimental design

Global versus local search in constrained optimization of computer models

Statistical Science

[Design and Analysis of Computer Experiments]: Rejoinder

Journal	Technometrics
Publisher	Informa UK Limited
ISSN	0040-1706
Open Access	No