Chapter 5 - Chapter 5 Local Surrogates 5.1 Introduction Polynomial response surface are primarily intended to model the overall behavior of

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
Chapter 5 Local Surrogates 5.1 Introduction Polynomial response surface are primarily intended to model the overall behavior of functions. If we have enough data to capture local waviness in a function, we can conceivably use high- order polynomials for that purpose. However, such polynomials have very poor extrapolation properties, so they are rarely used. Instead, we usually rely on surrogates that use local basis functions, () i x , instead or in addition to the global functions (such as monomials). In this chapter we will cover radial basis functions (RBF), kriging, and support vector regression (SVR). The assumption that local behavior can be captured is usually satisfied well only in the absence of noise. With substantial noise in the measurements, local variations are difficult to distinguish from noise. Therefore, most local surrogates tend to be used as interpolating surrogates, and one may get the impression that this is inherent in this surrogates, especially for kriging surrogates. However, they do have versions that are intended to be used with noisy data.
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 2
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 06/06/2011 for the course EAS 4240 taught by Professor Peterifju during the Spring '08 term at University of Florida.

Page1 / 4

Chapter 5 - Chapter 5 Local Surrogates 5.1 Introduction Polynomial response surface are primarily intended to model the overall behavior of

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online