Towards non-linear constraint estimation for expensive optimization

TY - GEN

T1 - Towards non-linear constraint estimation for expensive optimization

AU - Gieseke, Fabian

AU - Kramer, Oliver

N1 - Conference code: 16

PY - 2013

Y1 - 2013

N2 - Constraints can render a numerical optimization problem much more difficult to address. In many real-world optimization applications, however, such constraints are not explicitly given. Instead, one has access to some kind of a "black-box" that represents the (unknown) constraint function. Recently, we proposed a fast linear constraint estimator that was based on binary search. This paper extends these results by (a) providing an alternative scheme that resorts to the effective use of support vector machines and by (b) addressing the more general task of non-linear decision boundaries. In particular, we make use of active learning strategies from the field of machine learning to select reasonable training points for the recurrent application of the classifier. We compare both constraint estimation schemes on linear and non-linear constraint functions, and depict opportunities and pitfalls concerning the effective integration of such models into a global optimization process.

AB - Constraints can render a numerical optimization problem much more difficult to address. In many real-world optimization applications, however, such constraints are not explicitly given. Instead, one has access to some kind of a "black-box" that represents the (unknown) constraint function. Recently, we proposed a fast linear constraint estimator that was based on binary search. This paper extends these results by (a) providing an alternative scheme that resorts to the effective use of support vector machines and by (b) addressing the more general task of non-linear decision boundaries. In particular, we make use of active learning strategies from the field of machine learning to select reasonable training points for the recurrent application of the classifier. We compare both constraint estimation schemes on linear and non-linear constraint functions, and depict opportunities and pitfalls concerning the effective integration of such models into a global optimization process.

U2 - 10.1007/978-3-642-37192-9_46

DO - 10.1007/978-3-642-37192-9_46

M3 - Article in proceedings

SN - 978-3-642-37191-2

T3 - Lecture notes in computer science

SP - 459

EP - 468

BT - Applications of Evolutionary Computation

A2 - Esparcia-Alcázar, Anna I.

PB - Springer

T2 - 16th European Conference on Applications of Evolutionary Computation

Y2 - 3 April 2013 through 5 April 2013

ER -