SACOBRA (Self-adjusting Constrained Optimization By Radial Basis Functions) is a project which aims at solving constrained expensive problems with very few function evalutions. Such problems arise for example in car design and car manufacturing when complicated and time-consuming simulations have to check whether a new design fulfills all safety and stability constraints.

Our research uses surrogate modelling, mainly with the help of radial basis functions, but also with Kriging models, to allow optimization with very few function evaluations. Our real-valued and derivative-free optimizer takes special care to handle multiple constraints (equality or inequality) in parallel to the optimization goal.

 

 

 

SACOBRA is also a software package which is available for download here.

Application areas: optimization of wing profiles, crash simulations and weight optimization in automotive industries.

 

Project members

Fachhochschule Köln
© Schmülgen/TH Köln
Dr. Samineh Bagheri,
TH Köln
konen-3 Prof. Dr. Wolfgang Konen,
TH Köln

 

Honors

 

Publications

2020

Bagheri, Samineh

Self-Adjusting Surrogate-Assisted Optimization Techniques for Expensive Constrained Black Box Problems Promotionsarbeit

Leiden University and TH Köln, 2020, (PhD thesis).

BibTeX

2019

Bagheri, Samineh; Konen, Wolfgang; Bäck, Thomas

Solving Optimization Problems with High Conditioning by Means of Online Whitening Inproceedings

Lopez-Ibanez, Manuel (Hrsg.): Genetic and Evolutionary Computation Conference 2019 (GECCO'19), Prague, S. 243-244, ACM, 2019.

Links | BibTeX

Bagheri, Samineh; Konen, Wolfgang; Bäck, Thomas

SACOBRA with Online Whitening for Solving Optimization Problems with High Conditioning Forschungsbericht

arXiv preprint arXiv:1904.08397 2019.

Links | BibTeX

Bagheri, Samineh; Konen, Wolfgang; Bäck, Thomas

How to Solve the Dilemma of Margin-Based Equality Constraint Handling Methods Artikel

at-Automatisierungstechnik, submitted , 2019.

BibTeX

2018

Bagheri, Samineh; Konen, Wolfgang; Bäck, Thomas

How to Solve the Dilemma of Margin-Based Equality Handling Methods Inproceedings

Hoffmann, Frank; Hüllermeier, Eyke; Mikut, Ralf (Hrsg.): Proceedings - 28. Workshop Computational Intelligence, Dortmund, 29. - 30. November 2018, S. 257-270, KIT Scientific Publishing, Karlsruhe, 2018, ISBN: 978-3-7315-0845-8, (Young Author Award).

Links | BibTeX

2017

Bagheri, Samineh; Konen, Wolfgang; Bäck, Thomas

Comparing Kriging and Radial Basis Function Surrogates Inproceedings

Hoffmann, Frank; Hüllermeier, Eyke (Hrsg.): Proceedings 27. Workshop Computational Intelligence, S. 243-259, Universitätsverlag Karlsruhe, 2017.

Links | BibTeX

Bagheri, Samineh; Konen, Wolfgang; Allmendinger, Richard; Branke, Jürgen; Deb, Kalyanmoy; Fieldsend, Jonathan; Quagliarella, Domenico; Sindhya, Karthik

Constraint Handling in Efficient Global Optimization Inproceedings

Bosman, Peter A N (Hrsg.): Genetic and Evolutionary Computation Conference 2017 (GECCO'17), Berlin, S. 1, ACM, 2017.

Links | BibTeX

Bagheri, Samineh; Konen, Wolfgang; Emmerich, Michael; Bäck, Thomas

Self-adjusting parameter control for surrogate-assisted constrained optimization under limited budgets Artikel

Applied Soft Computing, 61 , S. 377–393, 2017, ISSN: 1568-4946.

Links | BibTeX

2016

Bagheri, Samineh; Konen, Wolfgang; Bäck, Thomas

Online Selection of Surrogate Models for Constrained Black-Box Optimization Inproceedings

Jin, Yaochu (Hrsg.): SSCI'2016, Athens, S. 1, IEEE, 2016, (Best Student Paper Award).

Links | BibTeX

Bagheri, Samineh; Konen, Wolfgang; Bäck, Thomas

Equality Constraint Handling for Surrogate-Assisted Constrained Optimization Inproceedings

Tan, Kay Chen (Hrsg.): WCCI'2016, Vancouver, S. 1, IEEE, 2016.

Links | BibTeX