-
![[PDF]](http://www.flowermountains.nl/wp-content/plugins/papercite/img/pdf.png)
D. Bloembergen, S. De Jong, and K. Tuyls, “Lenient Learning in a Multiplayer Stag Hunt,” in Proc. of 23rd Benelux Conf. on Artificial Intelligence (BNAIC 2011), 2011, pp. 44-50.
[Bibtex]
@inproceedings{Bloembergen2011a,
abstract = {This paper describes the learning dynamics of individual learners in a multiplayer Stag Hunt game, focussing primarily on the difference between lenient and non-lenient learning. We find that, as in 2-player games, leniency significantly promotes cooperative outcomes in 3-player games, as the basins of attraction of (partially) cooperative equilibria grow under this learning scheme. Moreover, we observe significant differences between purely selection-based models, as often encountered in related analytical research, and models that include mutation. Therefore, purely selection-based analysis might not always accurately predict the behavior of practical learning algorithms, which often include mutation.},
author = {Bloembergen, Daan and {De Jong}, Steven and Tuyls, Karl},
booktitle = {Proc. of 23rd Benelux Conf. on Artificial Intelligence (BNAIC 2011)},
pages = {44--50},
title = {Lenient Learning in a Multiplayer Stag Hunt},
pdf = {http://www.flowermountains.nl/pub/Bloembergen2011a.pdf},
year = {2011}
}
- D. Bloembergen, M. Kaisers, and K. Tuyls, “Empirical and Theoretical Support for Lenient Learning (Extended Abstract),” in Proc. of 10th Int. Conf. on Autonomous Agents and Multiagent Systems (AAMAS 2011), 2011, pp. 1105-1106.
[Bibtex]
@inproceedings{Bloembergen2011,
author = {Bloembergen, Daan and Kaisers, Michael and Tuyls, Karl},
booktitle = {Proc. of 10th Int. Conf. on Autonomous Agents and Multiagent Systems (AAMAS 2011)},
editor = {Tumer and Yolum and Sonenberg and Stone},
pages = {1105--1106},
publisher = {International Foundation for AAMAS},
title = {Empirical and Theoretical Support for Lenient Learning (Extended Abstract)},
pdf = {http://www.flowermountains.nl/pub/Bloembergen2011.pdf},
year = {2011}
}
- S. Alers, D. Bloembergen, D. Hennes, M. Bügler, and K. Tuyls, “MITRO: an augmented mobile telepresence robot with assisted control (Demo),” in Proc. of 9th European Workshop on Multi-agent Systems (EUMAS 2011), 2011.
[Bibtex]
@inproceedings{Alers2011b,
author = {Sjriek Alers and Daan Bloembergen and Daniel Hennes and Max B{\"{u}}gler and Karl Tuyls},
title = {MITRO: an augmented mobile telepresence robot with assisted control (Demo)},
booktitle = {Proc. of 9th European Workshop on Multi-agent Systems (EUMAS 2011)},
pdf = {http://www.flowermountains.nl/pub/Alers2011b.pdf},
year = {2011}
}
- S. Alers, D. Bloembergen, D. Hennes, and K. Tuyls, “Augmented mobile telepresence with assisted control (Demo),” in Proc. of 23rd Benelux Conf. on Artificial Intelligence (BNAIC 2011), 2011, pp. 451-452.
[Bibtex]
@inproceedings{Alers2011a,
author = {Alers, Sjriek and Bloembergen, Daan and Hennes, Daniel and Tuyls, Karl},
booktitle = {Proc. of 23rd Benelux Conf. on Artificial Intelligence (BNAIC 2011)},
pages = {451--452},
title = {Augmented mobile telepresence with assisted control (Demo)},
pdf = {http://www.flowermountains.nl/pub/Alers2011a.pdf},
year = {2011}
}
- S. Alers, D. Bloembergen, D. Hennes, S. De Jong, M. Kaisers, N. Lemmens, K. Tuyls, and G. Weiss, “Bee-inspired foraging in an embodied swarm (Demo),” in Proc. of 10th Int. Conf. on Autonomous Agents and Multiagent Systems (AAMAS 2011), 2011, pp. 1311-1312.
[Bibtex]
@inproceedings{Alers2011,
abstract = {We show the emergence of Swarm Intelligence in physical robots. We transfer an optimization algorithm which is based on bee-foraging behavior to a robotic swarm. In simulation this algorithm has already been shown to be more effective, scalable and adaptive than algorithms inspired by ant foraging. In addition to this advantage, bee-inspired foraging does not require (de-)centralized simulation of environmental parameters (e.g. pheromones).},
author = {Alers, Sjriek and Bloembergen, Daan and Hennes, Daniel and {De Jong}, Steven and
Kaisers, Michael and Lemmens, Nyree and Tuyls, Karl and Weiss, Gerhard},
booktitle = {Proc. of 10th Int. Conf. on Autonomous Agents and Multiagent Systems (AAMAS 2011)},
editor = {Tumer and Yolum and Sonenberg and Stone},
keywords = {foraging,swarm intelligence,swarm robotics},
pages = {1311--1312},
publisher = {International Foundation for AAMAS},
title = {Bee-inspired foraging in an embodied swarm (Demo)},
pdf = {http://www.flowermountains.nl/pub/Alers2011.pdf},
year = {2011}
}
- D. Bloembergen, M. Kaisers, and K. Tuyls, “A comparative study of multi-agent reinforcement learning dynamics,” in Proc. of 22nd Benelux Conf. on Artificial Intelligence (BNAIC 2010), 2010, pp. 11-18.
[Bibtex]
@inproceedings{Bloembergen2010a,
abstract = {Multi-agent learning plays an increasingly important role in solving complex dynamic problems in today’s society. Recently, an evolutionary game theoretic approach to multi-agent reinforcement learning has been proposed as a first step towards a more general theoretical framework. This article uses the evolutionary game theory perspective to link behavioral properties of learning algorithms to their performance in both homogeneous and heterogeneous games, thereby contributing to a better understanding of multi- agent reinforcement learning dynamics. Simulation experiments are performed in the domain of 2 × 2 normal form games with the learning algorithms Lenient and non-lenient Frequency Adjusted Q-learning, Finite Action-set Learning Automata and Polynomial Weights Regret Minimization. The results show that evolutionary game theory provides an efficient way to predict the behavior, convergence properties and performance of reinforcement learners. In general, leniency is found to be the preferable choice in cooperative games. Furthermore, the non-lenient learning algorithms do not show significant differences when their intrinsic learning speed is compensated for.},
author = {Bloembergen, Daan and Kaisers, Michael and Tuyls, Karl},
booktitle = {Proc. of 22nd Benelux Conf. on Artificial Intelligence (BNAIC 2010)},
pages = {11--18},
publisher = {University of Luxembourg},
title = {A comparative study of multi-agent reinforcement learning dynamics},
pdf = {http://www.flowermountains.nl/pub/Bloembergen2010a.pdf},
year = {2010}
}
- D. Bloembergen, M. Kaisers, and K. Tuyls, “Lenient frequency adjusted Q-learning,” in Proc. of 22nd Benelux Conf. on Artificial Intelligence (BNAIC 2010), 2010, pp. 19-26.
[Bibtex]
@inproceedings{Bloembergen2010,
abstract = {Overcoming convergence to suboptimal solutions in cooperative multi-agent games has been a main challenge in reinforcement learning. The concept of leniency has been proposed to be more forgiving for initial mis-coordination. It has been shown theoretically that an arbitrarily high certainty of convergence to the global optimum can be achieved by increasing the degree of leniency, but the relation of the evolutionary game theoretic model to the Lenient Q-learning algorithm relied on the simplifying assumption that all actions would be updated simultaneously. Building on insights from Frequency Adjusted Q-learning, this article introduces the variation Lenient Frequency Adjusted Q-learning that matches the theoretical model precisely, and allows for arbitrarily high convergence to Pareto optimal equilibria in cooperative games.},
author = {Bloembergen, Daan and Kaisers, Michael and Tuyls, Karl},
booktitle = {Proc. of 22nd Benelux Conf. on Artificial Intelligence (BNAIC 2010)},
pages = {19--26},
publisher = {University of Luxembourg},
title = {Lenient frequency adjusted Q-learning},
pdf = {http://www.flowermountains.nl/pub/Bloembergen2010.pdf},
year = {2010}
}
- D. Bloembergen, “Analyzing reinforcement learning algorithms using evolutionary game theory,” Master Thesis, 2010.
[Bibtex]
@mastersthesis{Bloembergen2010Thesis,
author = {Daan Bloembergen},
school = {Maastricht University},
title = {Analyzing reinforcement learning algorithms using evolutionary game theory},
year = {2010},
pdf = {http://www.flowermountains.nl/pub/Bloembergen2010Thesis.pdf}
}
