Publications

@inproceedings{Koch13a,

title = {Subsampling strategies in SVM ensembles},

author = {Patrick Koch and Wolfgang Konen},

editor = {Frank Hoffmann and Eyke Hüllermeier},

url = {http://www.gm.fh-koeln.de/~konen/Publikationen/kochGMA2013.pdf},

year  = {2013},

date = {2013-01-01},

booktitle = {Proceedings 23. Workshop Computational Intelligence},

pages = {119--134},

publisher = {Universitätsverlag Karlsruhe},

keywords = {SOMA, TDMR},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{stork13a,

title = {SVM ensembles are better when different kernel types are combined},

author = {Jörg Stork and Ricardo Ramos and Patrick Koch and Wolfgang Konen},

editor = {Berthold Lausen},

url = {http://www.gm.fh-koeln.de/~konen/Publikationen/storkECDA-2013.pdf},

year  = {2013},

date = {2013-01-01},

booktitle = {European Conference on Data Analysis (ECDA13)},

publisher = {(under review)},

keywords = {SOMA, TDMR},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{Koch12ab,

title = {Efficient sampling and handling of variance in tuning data mining models},

author = {Patrick Koch and Wolfgang Konen},

editor = {Carlos A. Coello Coello and Vincenzo Cutello and others},

url = {http://maanvs03.gm.fh-koeln.de/webpub/CIOPReports.d/Koch12a.d/Koch12a.pdf},

year  = {2012},

date = {2012-01-01},

booktitle = {PPSN'2012: 12th International Conference on Parallel Problem Solving From Nature, Taormina},

pages = {195--205},

publisher = {Springer},

address = {Heidelberg},

keywords = {SOMA, TDMR},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{Thil12b,

title = {Reinforcement learning with n-tuples on the game Connect-4},

author = {Markus Thill and Patrick Koch and Wolfgang Konen},

editor = {Carlos A. Coello Coello and Vincenzo Cutello},

url = {http://maanvs03.gm.fh-koeln.de/webpub/CIOPReports.d/Thi12.d/Thil12.pdf},

year  = {2012},

date = {2012-01-01},

booktitle = {PPSN'2012: 12th International Conference on Parallel Problem Solving From Nature, Taormina},

pages = {184--194},

publisher = {Springer},

address = {Heidelberg},

keywords = {SOMA},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{Zaef12a,

title = {Multi-Criteria Optimization for Hard Problems under Limited Budgets},

author = {Martin Zaefferer and Thomas Bartz-Beielstein and Martina Friese and Boris Naujoks and Oliver Flasch},

editor = {Terry Soule and Anne Auger and Jason Moore and David Pelta and Christine Solnon and Mike Preuss and Alan Dorin and Yew-Soon Ong and Christian Blum and Dario Landa Silva and Frank Neumann and Tina Yu and Aniko Ekart and Wil Browne and Tim Kovacs and Man-Leung Wong and Clara Pizzuti and Jon Rowe and Tobias Friedrich and Giovanni Squillero and Nicolas Bredeche and Stephen Smith and Alison Motsinger-Rei and Jose Lozano and Martin Pelikan and Silja Meyer-Nienber and Christian Igel and Greg Hornby and},

isbn = {978-1-4503-1178-6},

year  = {2012},

date = {2012-01-01},

booktitle = {GECCO Companion '12: Proceedings of the fourteenth international conference on Genetic and evolutionary computation conference companion},

pages = {1451--1452},

publisher = {ACM},

address = {Philadelphia, Pennsylvania, USA},

abstract = {Many relevant industrial optimization tasks feature more than just one quality criterion. State-of-the-art multi-criteria optimization algorithms require a relatively large number of function evaluations (usually more than 10^5) to approximate Pareto fronts. Due to high cost or time consumption this large amount of function evaluations is not always available. Therefore, it is obvious to combine techniques such as Sequential Parameter Optimization (SPO), which need a very small number of function evaluations only, with techniques from evolutionary multi-criteria optimization (EMO). In this paper, we show how EMO techniques can be efficiently integrated into the framework of the SPO Toolbox (SPOT). We discuss advantages of this approach in comparison to state-of-the-art optimizers. Moreover, with the resulting capability to allow competing objectives, the opportunity arises to not only aim for the best, but also for the most robust solution. Herein we present an approach to optimize not only the quality of the solution, but also its robustness, taking these two goals as objectives for multi-criteria optimization into account.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{Flas12b,

title = {Sequential Parameter Optimization for Symbolic Regression},

author = {Thomas Bartz-Beielstein and Oliver Flasch and Martin Zaefferer},

editor = {Steven Gustafson and Ekaterina Vladislavleva},

isbn = {978-1-4503-1178-6},

year  = {2012},

date = {2012-01-01},

booktitle = {GECCO 2012 Symbolic regression and modeling workshop},

pages = {495--496},

publisher = {ACM},

address = {Philadelphia, Pennsylvania, USA},

abstract = {Modern Symbolic Regression (SR) engines are complex systems of many components, most of which require some form of parameterization. In this talk, we show how to apply Sequential Parameter Optimization (SPO) as a rigorous method for finding near-optimal parameter settings for SR systems. As modern SR systems often offer alternative operator sets for population initialization, variation, and selection, we also demonstrate how to use modern Design of Experiments (DoE) methods to find problem-specific near-optimal SR system configurations, in addition to near-optimal parameterizations for each selected system component. The experimental design for SR can somehow be tricky, because of interactions in the parameter settings. Methods for handling configurations of parameters which depend on higher-level parameters will be presented. Our exposition is based on a simple framework for statistical sound, reproducible empirical research in SR.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{Bart12d,

title = {SPOT Applied to Non-Stochastic Optimization Problems---An Experimental Study},

author = {Thomas Bartz-Beielstein and Martina Friese and Boris Naujoks and Martin Zaefferer},

editor = {Katya Rodriguez and Christian Blum},

isbn = {978-1-4503-1178-6},

year  = {2012},

date = {2012-01-01},

booktitle = {GECCO 2012 Late breaking abstracts workshop},

pages = {645--646},

publisher = {ACM},

address = {Philadelphia, Pennsylvania, USA},

keywords = {SPOT},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{Bart12f,

title = {Statistical Analysis of Optimization Algorithms with R},

author = {Thomas Bartz-Beielstein and Mike Preuss and Martin Zaefferer},

editor = {Gabriela Ochoa},

isbn = {978-1-4503-1178-6},

year  = {2012},

date = {2012-01-01},

booktitle = {GECCO 2012 Specialized techniques and applications tutorials},

pages = {1259--1286},

publisher = {ACM},

address = {Philadelphia, Pennsylvania, USA},

abstract = {Based on experiences from several (rather theoretical) tutorials and workshops devoted to the experimental analysis of algorithms at the world's leading conferences in the field of Computational Intelligence, a practical, hands-on tutorial for the statistical analysis of optimization algorithms is presented. This tutorial -demonstrates how to analyze results from real experimental studies, e.g., experimental studies in EC -item gives a comprehensive introduction in the R language -item introduces the powerful GUI rstudio (http://rstudio.org) -exemplifies the analysis using SPOT (http://cran.r-project.org/web/packages/SPOT/) R is the most attractive and fastest growing open source computer language for statistical computing and graphics in the world. It provides a wide variety of statistical and graphical techniques: linear and nonlinear modeling, statistical tests, time series analysis, classification, clustering, etc. R is distributed over CRAN (http://cran.r-project.org), which is a network of ftp and web servers around the world that store identical, up-to-date, versions of code and documentation for R.},

keywords = {optimization},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{Bart11b,

title = {Noisy optimization with sequential parameter optimization and optimal computational budget allocation},

author = {Bartz-Beielstein, Thomas and Friese, Martina and Zaefferer, Martin and Naujoks, Boris and Flasch, Oliver and Konen, Wolfgang and Koch, Patrick},

url = {http://doi.acm.org/10.1145/2001858.2001926},

isbn = {978-1-4503-0690-4},

year  = {2011},

date = {2011-01-01},

booktitle = {Proceedings of the 13th annual conference companion on Genetic and evolutionary computation},

pages = {119--120},

publisher = {ACM},

address = {Dublin, Ireland},

series = {GECCO '11},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{Frie11ab,

title = {Ensemble Based Optimization and Tuning Algorithms},

author = {Martina Friese and Martin Zaefferer and Thomas Bartz-Beielstein and Oliver Flasch and Patrick Koch and Wolfgang Konen and Boris Naujoks},

editor = {F. Hoffmann and E. Hüllermeier},

year  = {2011},

date = {2011-01-01},

booktitle = {Proceedings 21. Workshop Computational Intelligence},

pages = {119--134},

publisher = {Universitätsverlag Karlsruhe},

keywords = {optimization},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{Frie11a,

title = {Ensemble Based Optimization and Tuning Algorithms},

author = {Friese, Martina and Zaefferer, Martin and Bartz-Beielstein, Thomasand Flasch, Oliver and Koch, Patrick and Konen, Wolfgang and Naujoks, Boris},

editor = {Hoffmann, Frank and Hüllermeier, Eyke},

year  = {2011},

date = {2011-01-01},

booktitle = {Proceedings 21. Workshop Computational Intelligence},

pages = {119--134},

publisher = {Universitätsverlag Karlsruhe},

keywords = {TDMR},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{Koch11bb,

title = {Tuned Data Mining in R},

author = {Patrick Koch and Wolfgang Konen and Boris Naujoks and Oliver Flasch and Martina Friese and Martin Zaefferer and Thomas Bartz-Beielstein},

editor = {Frank Hoffmann and Eyke Hüllermeier},

year  = {2011},

date = {2011-01-01},

booktitle = {Proceedings 21. Workshop Computational Intelligence},

pages = {147--160},

publisher = {Universitätsverlag Karlsruhe},

keywords = {Data Mining, SOMA, TDMR},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{Kone11db,

title = {Tuned Data Mining: A Benchmark Study on Different Tuners},

author = {Wolfgang Konen and Patrick Koch and Oliver Flasch and Thomas Bartz-Beielstein and Martina Friese and Boris Naujoks},

editor = {Natalio Krasnogor},

year  = {2011},

date = {2011-01-01},

booktitle = {GECCO '11: Proceedings of the 13th Annual Conference on Genetic andEvolutionary Computation},

pages = {1995--2002},

keywords = {Data Mining, SOMA, TDMR},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{Bart11g,

title = {Automatic and interactive tuning of algorithms},

author = {Thomas Bartz-Beielstein and Mike Preuss},

url = {http://doi.acm.org/10.1145/2001858.2002141},

year  = {2011},

date = {2011-01-01},

booktitle = {GECCO (Companion)},

pages = {1361-1380},

crossref = {Kras11a},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{Bart10p,

title = {SPOT: A Toolbox for Interactive and Automatic Tuning in the R Environment},

author = {Thomas Bartz-Beielstein and Oliver Flasch and Patrick Koch and Wolfgang Konen},

editor = {Frank Hoffmann and Eyke Hüllermeier},

year  = {2010},

date = {2010-01-01},

booktitle = {Proceedings 20. Workshop Computational Intelligence},

pages = {264--273},

publisher = {Universitätsverlag Karlsruhe},

keywords = {SPOT},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{Flas10eb,

title = {Comparing SPO-tuned GP and NARX prediction models for stormwater tank fill level prediction},

author = {Oliver Flasch and Thomas Bartz-Beielstein and Arthur Davtyan and Patrick Koch and Wolfgang Konen and Tosin Oyetoyan and Michael Tamutan},

editor = {Gary Fogel and et al.},

year  = {2010},

date = {2010-01-01},

booktitle = {Proc. IEEE Congress Evolutionary Computation (CEC)},

pages = {1579--1586},

keywords = {GP, Prediction},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{Thill20a,

title = {Time Series Encodings with Temporal Convolutional Networks},

author = {Markus Thill and Wolfgang Konen and Thomas Bäck},

editor = {Bogdan Filipic Massimiliano Vasile},

url = {http://www.gm.fh-koeln.de/ciopwebpub/Thill20a.d/bioma2020-tcn.pdf},

year  = {2020},

date = {2020-01-01},

booktitle = {9th International Conference on Bioinspired Optimisation Methods and Their Applications (BIOMA)},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{Konen20bc,

title = {Reinforcement Learning for N-Player Games: The Importance of Final Adaptation},

author = {Wolfgang Konen and Samineh Bagheri},

editor = {Bogdan Filipic Massimiliano Vasile},

url = {http://www.gm.fh-koeln.de/ciopwebpub/Konen20b.d/bioma20-TDNTuple.pdf},

year  = {2020},

date = {2020-01-01},

booktitle = {9th International Conference on Bioinspired Optimisation Methods and Their Applications (BIOMA)},

address = {Bruxelles},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{Konen19b,

title = {General Board Game Playing for Education and Research in Generic AI Game Learning},

author = {Wolfgang Konen},

editor = {Diego Perez and Sanaz Mostaghim and Simon Lucas},

url = {https://arxiv.org/pdf/1907.06508},

year  = {2019},

date = {2019-08-20},

booktitle = {IEEE Conference on Games},

address = {London},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{Bagh19c,

title = {Solving Optimization Problems with High Conditioning by Means of Online Whitening},

author = {Samineh Bagheri and Wolfgang Konen and Thomas Bäck},

editor = {Manuel Lopez-Ibanez},

url = {http://blogs.gm.fh-koeln.de/ciop/files/2019/07/GECCO2019.pdf},

year  = {2019},

date = {2019-07-01},

booktitle = {Genetic and Evolutionary Computation Conference 2019 (GECCO'19), Prague},

pages = {243-244},

publisher = {ACM},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}