GBG (General Board Game Playing & Learning) is an Open Source software framework developed at Technical University of Cologne. GBG aims to ease the entry for the students in Game learning research area which is a very interesting sub-field of artifical intelligence. Last year, two students finished their Bachelor theses successfully on the basis of the GBG under supervision of Prof. Konen. 

GBG framework can already be used for research purposes by comparing performances of different implemented intelligent agents on different implemented games. But the most interesting fact about GBG is, that any other board game or agent can easily be embedded and used. If you are interested you can read more about GBG here.

Thanks to the flexibility of GBG, many new themes in Game learning and AI can be considered as a possible Bachelor or Master thesis topics. The topics can be either implementation of a new game and analysing the performance of the currently implemented agents on such games or implementing a new agent (learning algorithm) and study its effectiveness on the already existing games. 

 In case you are interested in any of the following topics (or anythign similar) please contact Prof. Konen.

Implementing:

Research

  • Hex in GBG
  • Dots and boxes in GBG
  • ...

Software engineeing

  • Client-Server-Architektur für GBG
  • Competition Framework in GBG (Round-robin Tournaments, Multi-objective Visualization)
  • ...

New ideas are also appreciated.

The annual Computational Intelligence workshop in Dortmund is a great place for sharing new ideas and recent work done in computational-intelligence related topics. This year, the GMA workshop took place on 23-24 of November at the Dorint Hotel in Dortmund. Our team presented two papers in the fields of surrogate models and time series analysis. In total there were 4 papers presented from members of TH Köln, Campus Gummersbach.

Besides receiving effective feedbacks about our own work, we also had a chance of  listening to many interesting talks about related research work, done in different universities. The presentations had very diverse topics: from more practical ones to very abstract and mathematical ones. The friendly atmosphere, constructive critics and open discussions make this workshop very special and productive.

 

If you are interested in our recent article Self-adjusting parameter control for surrogate-assisted constrained optimization under limited budgets  which appeared in the journal Applied Soft Computing (ASOC), please take the opportunity to follow this link: https://authors.elsevier.com/a/1Vj555aecSVnxb. For a period until November, 02, 2017, this link will provide the free download of the final journal article. (After this period, the link is still

(read more...)

 

The OC Digital Day organiezed by TH Köln and Opitz Consulting group has taken place at TH Köln Campus Gummersbach last Sturday. Many talks and a few keynotes were presented in the field of digitalising (model-based optimization, Big Data, Oracle, Gamification, smart home, ...) by Opitz employees as well as TH Köln students. Our group also joined this coneference with a presentation about the main ideas behind the model-based optimization for real world problems.

You can find a short video about Digital Day here.

We had the chance to present our work in the field of online model selection for surrogate assisted multivariate optimization in the 15th workshop of quality improvment methods which was held in Dortmund and organized by the statistics faculty of TU Dortmund. You can read more about the quality improvemnet annual workshops here.