Thomas Bartz-Beielstein
Research interests
-
InterestsSix Sigma, Evolutionary Algorithms, Computational Intelligence, Optimization, Simulation and Optimization, Sequential Parameter Optimization, Design and Analysis of Experiments, Parameter Tuning, Design and Analysis of Computer Experiments, Computational Science, Optimization Methods
Publications
-
Tuned data mining: a benchmark study on different tuners.
13th Annual Genetic and Evolutionary Computation Conference, GECCO 2011, Proceedings, Dublin, Ireland, July 12-16, 2011; 01/2011
-
Noisy optimization with sequential parameter optimization and optimal computational budget allocation
Proceedings of the 13th annual conference companion on Genetic and evolutionary computation, New York, NY, USA; 01/2011
-
Ensemble Based Optimization and Tuning Algorithms
Proceedings 21. Workshop Computational Intelligence; 01/2011
-
Sequential Parameter Optimization and Optimal Computational Budget Allocation for Noisy Optimization Problems
CIOP Reports. 01/2011;
Sequential parameter optimization (SPO) is a heuristic that combines classical and modern statistical techniques to improve the performance of search algorithms. It includes a broad vari- ety of meta models, e.g., linear models, random forest, and Gaussian process models (Kriging). The selection of ... [more] Sequential parameter optimization (SPO) is a heuristic that combines classical and modern statistical techniques to improve the performance of search algorithms. It includes a broad vari- ety of meta models, e.g., linear models, random forest, and Gaussian process models (Kriging). The selection of an adequate meta model can have signi�cant impact on SPO's performance. A comparison of di�erent meta models is of great importance. A recent study indicated that random forest based meta models might be a good choice. This rather surprising result will be analyzed in this paper. Moreover, Optimal Computing Budget Allocation (OCBA), which is an enhanced method for handling the computational budget spent for selecting new design points, is presented. The OCBA approach can intelligently determine the most e�cient replication numbers. We propose the integration of OCBA into SPO. In this study, SPO is directly used as an optimization method on di�erent noisy mathemat- ical test functions. This is di�ers from the standard way of using SPO for tuning algorithm parameters in the context of complex real-world applications. Using SPO this way allows for a comparison to other optimization algorithms. Our results reveal that the incorporation of OCBA and the selection of Gaussian pro- cess models are highly bene�cial. Moreover, SPO outperformed three di�erent alternative optimization algorithms on a set of �ve noisy mathematical test functions.
-
Automatic and interactive tuning of algorithms.
13th Annual Genetic and Evolutionary Computation Conference, GECCO 2011, Companion Material Proceedings, Dublin, Ireland, July 12-16, 2011; 01/2011
-
SPOT: An R Package For Automatic and Interactive Tuning of Optimization Algorithms by Sequential Parameter Optimization
06/2010;
The sequential parameter optimization (SPOT) package for R is a toolbox for tuning and understanding simulation and optimization algorithms. Model-based investigations are common approaches in simulation and optimization. Sequential parameter optimization has been developed, because there is a stron... [more] The sequential parameter optimization (SPOT) package for R is a toolbox for tuning and understanding simulation and optimization algorithms. Model-based investigations are common approaches in simulation and optimization. Sequential parameter optimization has been developed, because there is a strong need for sound statistical analysis of simulation and optimization algorithms. SPOT includes methods for tuning based on classical regression and analysis of variance techniques; tree-based models such as CART and random forest; Gaussian process models (Kriging), and combinations of different meta-modeling approaches. This article exemplifies how SPOT can be used for automatic and interactive tuning. Comment: Related software can be downloaded from http://cran.r-project.org/web/packages/SPOT/index.html
-
Clustering Based Niching for Genetic Programming in the R Environment
Proceedings 20. Workshop Computational Intelligence; 01/2010
-
Parameter-Tuned Data Mining: A General Framework
Proceedings 20. Workshop Computational Intelligence; 01/2010
-
Tuning and experimental analysis in evolutionary computation: what we still have wrong.
Genetic and Evolutionary Computation Conference, GECCO 2010, Proceedings, Portland, Oregon, USA, July 7-11, 2010, Companion Material; 01/2010
Following (59)
-
Matteo Gagliolo
Université Libre de Bruxelles -
Mátyás Brendel
Inventeurs du monde numérique -
Mike Preuss
Technische Universität Dortmund -
Jano van Hemert
Optos -
Álvaro Fialho
GE Global Research Brazil
Topics (4)
