Home
What is the SURROGATES toolbox?
The SURROGATES toolbox is a general-purpose library of multidimensional function approximation and optimization methods for MATLAB and OCTAVE. The current version includes, among others, the following capabilities:
Surrogates: Gaussian process, kriging, polynomial response surface, radial basis neural network, and linear Shepard.
Classical error analysis.
Variants of the efficient global optimization (EGO) algorithm
The SURROGATES toolbox uses the following collection of third party software:
GPML by Rasmussen and Williams [1]: www.gaussianprocess.org/gpml
DACE by Lophaven et al. [2]: www2.imm.dtu.dk/~hbn/dace
RBF by Jekabsons [3]: www.cs.rtu.lv/jekabsons/regression
SVM by Gunn [4]: www.isis.ecs.soton.ac.uk/resources/svminfo
Returning the favor
I strongly ask the user's community to give credit to the individual components and to the SURROGATES toolbox in any publication derived from the use of the toolbox. For example, when I publish my papers I usually have a paragraph like this:
Table 1 details the different surrogates used during this investigation. The SURROGATES toolbox was also used for easy manipulation of the surrogates.
Table 1: Setup for the set of used surrogates. The GPML[1], DACE [2], MATLAB neural networks [5], RBF [3], SURROGATES [7], and SVM [4] toolboxes were used to run the Gaussian process, kriging, radial basis neural network, radial basis function, linear Shepard algorithms, and support vector regression algorithms, respectively.
Download
References
[1] CE Rasmussen and CK Williams, Gaussian Processes for Machine Learning, The MIT Press, 2006.
[2] SN Lophaven, HB Nielsen, and J Sondergaard, "DACE - a MATLAB kriging toolbox," Tech. Rep. IMM-TR-2002-12, Technical University of Denmark, Denmark, Aug 2002, available at http://www2.imm.dtu.dk/~hbn/dace/.
[3] G Jekabsons, RBF: Radial Basis Function interpolation for MATLAB/OCTAVE, Riga Technical University, Latvia, version 1.1 ed., 2009, available at http://www.cs.rtu.lv/jekabsons/regression.html.
[4] SR Gunn, "Support vector machines for classification and regression," Tech. Rep., University of Southampton, UK, 1997, available at http://www.isis.ecs.soton.ac.uk/resources/svminfo/.
[5] MathWorks contributors, MATLAB The language of technical computing, The MathWorks, Inc, Natick, MA, USA, version 7.0 release 14 ed., 2004.
[6] WI Thacker, J Zhang, LT Watson, JB Birch, MA Iyer, and MW Berry, "Algorithm 905: SHEPPACK: modified Shepard algorithm for interpolation of scattered multivariate data," ACM Transactions on Mathematical Software, Vol. 37, No. 3, 2010, pp. 1-20.
[7] FAC Viana, SURROGATES Toolbox User's Guide, Gainesville, FL, USA, version 3.0 ed., 2011, available at https://sites.google.com/site/srgtstoolbox/.