The construction of surrogate models, such as radial basis function (RBF) and Kriging-based surrogates, requires an invertible (square and full rank matrix) or pseudoinvertible (overdetermined) linear system to be solved. This study demonstrates that the method used to solve this linear system may result in up to five orders of magnitude difference in the accuracy of the constructed surrogate model using exactly the same information. Hence, this paper makes the canonic and important point toward reproducible science: the details of solving the linear system when constructing a surrogate model must be communicated. This point is clearly illustrated on a single function, namely the Styblinski–Tang test function by constructing over 200 RBF surrogate models from 128 Latin Hypercubed sampled points. The linear system in the construction of each surrogate model was solved using LU, QR, Cholesky, Singular-Value Decomposition, and the Moore–Penrose pseudoinverse. As we show, the decomposition method influences the utility of the surrogate model, which depends on the application, i.e., whether an accurate approximation of a surrogate is required or whether the ability to optimize the surrogate and capture the optimal design is pertinent. Evidently the selection of the optimal hyperparameters based on the cross validation error also significantly impacts the utility of the constructed surrogate. For our problem, it turns out that selecting the hyperparameters at the lowest cross validation error favors function approximation but adversely affects the ability to optimize the surrogate model. This is demonstrated by optimizing each constructed surrogate model from 16 fixed initial starting points and recording the optimal designs. For our problem, selecting the optimal hyperparameter that coincides with the lowest monotonically decreasing function value significantly improves the ability to optimize the surrogate for most solution strategies.

References

References
1.
Powell
,
M. J. D.
,
2002
, “
Radial Basis Function Methods for Interpolation to Functions of Many Variables
,”
Fifth Hellenic-European Conference on Computer Mathematicas and its Applications
,
Athens
.
2.
Micchelli
,
C. A.
,
1986
, “
Interpolation of Scattered Data Distance Matrices and Conditionally Positive Definite Functions
,”
Constr. Approx.
,
2
, pp.
11
22
.
3.
Broomhead
,
D. S.
, and
Lowe
,
D.
,
1988
, “
Multivariable Functional Interpolation and Adaptive Networks
,”
Complex Syst.
,
2
, pp.
321
355
.
4.
Orr
,
M. J.
,
1996
, “
Introduction to Radial Basis Function Networks
,”
Center for Cognitive Science, University of Edinburgh
,
Technical Report
.
5.
Qian
,
Z.
,
Seepersad
,
C. C.
,
Joseph
,
V. R.
,
Allen
,
J. K.
, and
Wu
,
C. F. J.
,
2006
, “
Building Surrogate Models Based on Detailed and Approximate Simulations
,”
J. Mech. Design
,
128
(
4
), pp.
668
677
.
6.
Yu
,
K.
,
Yang
,
X.
, and
Mo
,
Z.
,
2014
, “
Profile Design and Multifidelity Optimization of Solid Rocket Motor Nozzle
,”
J. Fluids Eng.
,
136
(
3
),
031104
.
7.
Hu
,
Z.
, and
Du
,
X.
,
2015
, “
Mixed Efficient Global Optimization for Time-Dependent Reliability Analysis
,”
J. Mech. Design
,
137
(
5
),
051401
.
8.
Karkevandi-Talkhooncheh
,
A.
,
Rostami
,
A.
,
Hemmati-Sarapardeh
,
A.
,
Ahmadi
,
M.
,
Husein
,
M. M.
, and
Dabir
,
B.
,
2018
, “
Modeling Minimum Miscibility Pressure During Pure and Impure CO2 Flooding Using Hybrid of Radial Basis Function Neural Network and Evolutionary Techniques
,”
Fuel
,
220
, pp.
270
282
.
9.
Aljarah
,
I.
,
Faris
,
H.
,
Mirjalili
,
S.
, and
Al-Madi
,
N.
,
2018
, “
Training Radial Basis Function Networks Using Biogeography-Based Optimizer
,”
Neural Comput. Appl.
,
29
(
7
), pp.
529
553
.
10.
Palm
,
E.
,
Schwenker
,
F.
, and
Kestler
,
H.
,
2001
, “
Three Learning Phases for Radial Basis Function Networks
,”
Neural Netw.
,
14
, pp.
439
458
.
11.
Kumar
,
R. H.
,
Kumar
,
B. V.
,
Karthik
,
K.
,
Chand
,
J. L. K.
, and
Kumar
,
C. N.
,
2012
, “
Performance Analysis of Singular Value Decomposition (SVD) and Radial Basis Function (RBF) Neural Networks for Epilepsy Risk Levels Classifications From EEG Signals
,”
ResearchGate
., p.
5
.
12.
Nemati
,
K.
,
2018
, “
Massive Training Radial Basis Function Neural Network for Distinguishing Between Nodule and Non-Nodule
,”
J. Adv. Comput. Res.
,
9
(
2
), pp.
49
70
.
13.
Ferreira
,
A.J.M.
,
Kansa
,
E.J.
,
Fasshauer
,
G.E.
, and
Leitao
,
V.M.A.
,
2009
,
Progress on Meshless Methods
,
Springer Science and Business Media
,
New York
, pp.
57
75
.
14.
Kindelan
,
M.
, and
Moscoso
,
M.
,
2016
, “
Radial Basis Function Interpolation in the Limit of Increasingly Flat Basis Functions
,”
J. Comput. Phys.
,
307
, pp.
225
242
.
15.
Schaback
,
R.
,
1994
, “
Lower Bounds for Norms of Inverses of Interpolation Matrices for Radial Basis Functions
,”
J. Approx. Theor.
,
79
(
2
), pp.
287
306
.
16.
Schaback
,
R.
,
2008
, “
Limit Problems for Interpolation by Analytic Radial Basis Functions
,”
J. Comput. Appl. Math.
,
212
, pp.
127
149
.
17.
Buhmann
,
M. D.
,
Marchi
,
S. D.
, and
Perracchione
,
E.
,
2018
, “
Analysis of a New Class of Rational RBF Expansions
,”
IMA J. Numer. Anal
, pp.
22
.
18.
Wang
,
H.
,
Gu
,
J.
,
Di
,
X.
,
Liu
,
D.
,
Zhao
,
J.
, and
Sui
,
X.
,
2017
, “
Research on Classification and Recognition of Attacking Factors Based on Radial Basis Function Neural Network
,”
Cluster. Comput.
20
, pp.
1
13
. 10.1016/j.enzmictec.2008.09.015
19.
Liu
,
L.
,
Sun
,
J.
,
Zhang
,
D.
,
Du
,
G.
,
Chen
,
J.
, and
Xu
,
W.
,
2009
, “
Culture Conditions Optimization of Hyaluronic Acid Production by Streptococcus Zooepidemicus Based on Radial Basis Function Neural Network and Quantum-Behaved Particle Swarm Optimization Algorithm
,”
Enzyme. Microb. Technol.
,
44
(
1
), pp.
24
32
.
20.
Mohammadi
,
M.
,
Krishna
,
A.
,
Nalesh
,
S.
, and
Nandy
,
S. K.
,
2018
, “
A Hardware Architecture for Radial Basis Function Neural Network Classifier
,”
IEEE. Trans. Parallel Distrib. Syst.
,
29
(
3
), pp.
481
495
.
21.
Wang
,
X.
,
Liu
,
Y.
,
Sun
,
W.
,
Song
,
X.
, and
Zhang
,
J.
,
2018
, “
Multidisciplinary and Multifidelity Design Optimization of Electric Vehicle Battery Thermal Management System
,”
J. Mech. Design
,
140
(
9
),
094501
.
22.
Ball
,
K.
,
Sivakumar
,
N.
, and
Ward
,
J. D.
,
1992
, “
On the Sensitivity of Radial Basis Interpolation to Minimal Data Separation Distance
,”
Constr. Approx.
,
8
(
4
), pp.
401
426
.
23.
Narcowich
,
F. J.
,
Sivakumar
,
N.
, and
Ward
,
J. D.
,
1994
, “
On Condition Numbers Associated with Radial-Function Interpolation
,”
J. Math. Anal. Appl.
,
186
, pp.
457
485
.
24.
Narcowich
,
F. J.
, and
Ward
,
J. D.
,
1991
, “
Norms of Inverses and Condition Numbers for Matrices Associated With Scattered Data
,”
J. Approx. Theory
,
64
(
1
), pp.
69
94
.
You do not currently have access to this content.