Category Archives: Publications

Free Download: Proceedings 26th Workshop Computational Intelligence 2016

Die Proceedings zum 26. Workshop Computational Intelligence sind beim Verlag Scientific Publishing als Buch oder als PDF zum freien Download  erhältlich.
Vom SPOTSeven Team wurden die folgenden Beiträge veröffentlicht:

  • S. Chandrasekaran, M. Zaefferer, S. Moritz, J. Stork, M. Friese, A. Fischbach, T. Bartz-Beielstein: Data Preprocessing: A New Algorithm for Univariate Imputation Designed Specically for Industrial Needs und
  • A. Fischbach, M. Zaefferer, J. Stork, M. Friese, T. Bartz-Beielstein: From Real World Data to Test Functions.

Weitere Infos zum CI Workshop sind unter http://www.rst.e-technik.tu-dortmund.de/cms/de/Veranstaltungen/GMA-Fachausschuss/ zu finden.

Free Preprint: Model-based Methods for Continuous and Discrete Global Optimization

The paper “Model-based Methods for Continuous and Discrete Global Optimization” (T. Bartz-Beielstein, M. Zaefferer) is available for download via Cologne Open Science: https://cos.bibl.th-koeln.de/frontdoor/index/index/docId/435.
Surrogate models gained importance for discrete optimization problems. This paper presents a survey of model-based methods, focusing on continuous optimization. It introduces a taxonomy, which is useful as a guideline for selecting adequate model-based optimization tools, and provides details for the case of discrete optimization problems. Six strategies for dealing with discrete data structures are introduced.
A new approach for combining surrogate information via stacking is proposed. The implementation of this approach will be available in the open source R package SPOT2. The article concludes with a discussion of recent developments and challenges in both application domains. Continue reading

New Paper: A Surrogate Model Assisted EA for Computationally Expensive Design Optimization Problems with Discrete Variables

The paper “A surrogate model assisted evolutionary algorithm for computationally expensive design optimization problems with discrete variables” focuses on discrete numerical variables, which are sometimes unavoidable in product design and manufacturing. It was written by Bo Liu (Department of Computing, Glyndwr University, UK), Nan Sun (Department of Engineering, Glyndwr University, UK),  Qingfu Zhang (Department of Computer Science, City University of Hong Kong, Kowloon, Hong Kong SAR), Vic Grout (Department of Computing, Glyndwr University, UK), and Georges Gielen (ESAT-MICAS, Katholieke Universiteit Leuven, Belgium). The paper can be  accessed here: http://ieeexplore.ieee.org/abstract/document/7743986/

Related papers, e.g., M. Zaefferer, J. Stork, and T. Bartz-Beielstein, “Distance Measures for Permutations in Combinatorial Efficient Global Optimization” can be found on the SPOTSeven publication page http://www.spotseven.de/publications/.

GPU-based parallel evolution strategy in bioinformatics

The paper “GPU-based Point Cloud Superpositioning for Structural Comparisons of Protein Binding Sites” (Matthias Leinweber, Thomas Fober, and Bernd Freisleben) presents “a novel approach to solve the labeled point cloud superpositioning problem for performing structural comparisons of protein binding sites. The solution is based on a parallel evolution strategy that operates on large populations and runs on GPU hardware. The proposed evolution strategy reduces the likelihood of getting stuck in a local optimum of the multimodal real-valued optimization problem represented by labeled point cloud superpositioning.”
The authors applied the SPOT (sequential parameter optimization toolbox) to tune the evolution strategy: “The ES parameters population size, lifetime of individuals, recombination parameter, and mutation rate depend on the problem to be solved, and we used another optimizer, the Sequential Parameter Optimization Toolbox (SPOT) [53], to optimize them.”

The paper can be downloaded from http://doi.ieeecomputersociety.org/10.1109/TCBB.2016.2625793

References Continue reading

26. Workshop Computational Intelligence: Proceedings vom Vorjahr zum freien Download erhältlich

ci2016

Das SPOTSeven Lab ist mit zwei Beiträgen bei dem diesjährigen CI Workshop, 24.-25. November 2016 in Dortmund, vertreten:

  • A. Fischbach, J. Stork, M. Zaefferer, S. Krey, T. Bartz-Beielstein, Analyzing Capabilities of Latin Hypercube Designs Compared to Classical Experimental Design Methods
  • Sowmya Chandrasekaran, Steffen Moritz, Martin Zaefferer, Jörg Stork, Thomas Bartz-Beielstein, Data Preprocessing: A New Algorithm for Univariate Imputation Designed Specifically for Industrial Needs

Die Proceedings zum 25. Workshop Computational Intelligence sind beim Verlag Scientific Publishing als Buch oder als PDF zum freien Download  erhältlich.