Sprungmarken

Servicenavigation

Hauptnavigation

Sie sind hier:

Hauptinhalt

IRF-IT

Dipl.-Inf. Markus Kemmerling Foto von Dipl.-Inf. Markus Kemmerling

Adresse  

Technische Universität Dortmund
Institut für Roboterforschung
Abteilung Informationstechnik
Otto-Hahn-Str. 8, Einfahrt 32-34
44221 Dortmund

Info

Publications

2013
Grimme, Christian, Markus Kemmerling and Joachim Lepping: On the integration of theoretical single-objective scheduling results for multi-objective problems. In EVOLVE - A bridge between Probability, Set Oriented Numerics and Evolutionary Computation, Studies in Computational Intelligence, vol. 447, p. 333-363, Springer, 2013 [link]
2012
Kemmerling, Markus, Niels Ackermann and Mike Preuss: Making diplomacy bots individual. In Believable Bots, Editor: Philip F. Hingston, p. 265-288, Springer, 2012 [link]
2011
Grimme, Christian, Markus Kemmerling and Joachim Lepping: An Expertise-Guided Multi-Criteria Approach to Scheduling Problems. In Prococeedings of the Genetic and Evolutionary Computation Conference (GECCO 2011), p. 47-48, ACM, 2011 [link]
Kemmerling, Markus, Niels Ackermann and Mike Preuss: Nested Look-Ahead Evolutionary Algorithm Based Planning for a Believable Diplomacy Bot. In Applications of Evolutionary Computation - EvoApplications 2011, Part I, Lecture Notes in Computer Science (LNCS), vol. 6624, p. 83-92, Springer, 2011 [link]
2010
Kemmerling, Markus and Mike Preuss: Automatic adaptation to generated content via car setup optimization in TORCS. In Proceedings of the IEEE Conference on Computational Intelligence and Games (CIG 2010), p. 131-138, IEEE Press, 2010. (best paper nomination) [link]
2009
Kemmerling, Markus, Niels Ackermann, Nicola Beume, Mike Preuss, Sebastian Uellenbeck and Wolfgang Walz: Is human-like and well playing contradictory for diplomacy bots?. In Proceedings of the IEEE Symposium on Computational Intelligence and Games (CIG 2009), p. 209-216, IEEE Press, 2009 [link]
2008
Ackermann, Niels, Alireza Gholaman, Matthias Grochowski, Marc Gorzala, Markus Kemmerling, Daniel Spierling, Thomas Harweg, Sebastian Uellenbeck and Wolfgang Walz: Emotionally affected poker players. Demonstration Paper. Workshop within 10th International Conference on Parallel Problem Solving From Nature (PPSN 2008), 2008 [link]

 

Teaching

Summer Term 2012 Tutorial/Lab Course: Scheduling Problems - Algorithms and Applications (English)
Winter Term 2011/2012 Seminar: Optimization Methods II (German)
Summer Term 2011 Tutorial/Lab Course: Scheduling Problems - Algorithms and Applications (German)
Winter Term 2010/2011 Project Group: Multiobjective Scheduling (English)

 

Research Projects

2010 - 2012 BMBF Project C3Grid-INAD: towards an INfrastructure for general Access to climate Data
2010 Diploma Thesis Car setup optimization based on noisy data from a car racing simulator
    Original title: Optimierung der Fahrzeugabstimmung auf Basis verrauschter Daten einer Autorennsimulation [pdf]
2008 - 2009 Project Group Modelling human-like non-player characters for strategy games using computational intelligence techniques

 

Reviewing Activities

2013
IEEE IPDPS 2013
2012
EURO-PAR 2012
GECCO 2012 - Digital Entertainment Technologies and Arts Track
2011
IEEE Cloudcom 2011
4. DFN-Forum Kommunikationstechnologien
GECCO 2011 - Digital Entertainment Technologies and Arts Track

 

Program Committee Activities

Digital Entertainment Technologies and Arts Track, GECCO 2012
Digital Entertainment Technologies and Arts Track, GECCO 2011

 

Organizing Committee Activities

Car Setup Optimization Competition, EvoStar 2010

 

Software

The Stragotiator

The Stragotiator is an Artificial Intelligence (AI), also called Non-Player Character (NPC) or bot, for the game Diplomacy. He is implemented in Java and designed to be used with the Diplomacy AI Development Environment (DAIDE).

In contrast to other existing bots, the intended goal was the development of a human-like bot that is fun to play against. While most bots focusing only on playing strength and how to win, the greatest efforts in the Stragotiator project were done to create a bot with good negotiation capabilities that interacts with other players and appears to be mostly human. This efforts resulted in a bot capable of press level 20 of the DAIDE language. An adequate amount of playing strength was achieved with elaborate heuristics and evolutionary movement optimization. Only convoy moves and demilitarized zones are not yet support by the Stragotiator. Since the most time while playing the board game is spend in negotiations, the Stragotiator is the best choice for all who want to play against a bot. He plays quite well, is a reliable ally and supporter and an intelligent negotiator.

The zip file contains the Stragotiator in binary form together with all sources and documentations.

Stragotiator.zip

 

Noise Measurement

Java implementation of an uncertainty measurement according to Hansen, Nikolaus, Andre P. N. Niederberger, Lino Guzzella and Petros Koumoutsakos: A method for handling uncertainty in evolutionary optimization with an application to feedback control of combustion. IEEE Transactions on Evolutionary Computation, 13(1):180-197, 2009

This is a porting of the function local_noisemeasurement in Nikolaus Hansen's Matlab implementation of the Covariance Matrix Adaptation Evolution Strategy (CMA-ES) Version 3.40.beta [link].

The zip file contains the noise measurement as Java source code, JUnit tests, and a short example how to use the noise measurement together with the CMA-ES implementation of Nikolaus Hansen.

NoiseMeasurement.zip