Document Actions

Publications

	You may also check out my new Google Scholar profile! 2012 M. Tokic, P. Ertle, G. Palm, D. Söffker, H. Voos. Robust Exploration/Exploitation Trade-Offs in Safety-Critical Applications. In Proceedings of the 8th IFAC International Symposium on Fault Detection, Supervision and Safety of Technical Processes (SafeProcess 2012). (to appear). P. Ertle, M. Tokic, T. Bystricky, M. Ebel, H. Voos and D. Söffker. Conceptual Design of a Dynamic Risk-Assessment Server for Autonomous Robots. In Proceedings of the 7th German Conference on Robotics (Robotik 2012). (to appear). 2011 M. Tokic and G. Palm. Value-Difference Based Exploration: Adaptive Control Between Epsilon-Greedy and Softmax. In J. Bach & S. Edelkamp (Eds.), KI 2011: Advances in Artificial Intelligence, pp. 335-346. Springer Verlag, Heidelberg, Germany, 2011. [ bib ] [ pdf, the original publication is available at www.springerlink.com ] S. Montresor, J. Kay, M. Tokic and J. Summerton. Work In Progress: Programming in a Confined Space - A Case Study in Porting Modern Robot Software to an Antique Platform. In Proceedings of the 41st ASEE/IEEE Frontiers in Education Conference, pp. T3H-1 - T3H-3. IEEE Press, USA, 2011. [ bib] [ pdf, the original publication is available at ieeexplore.ieee.org ] 2010 M. Tokic, A. Usadel, J. Fessler, and W. Ertel. On an educational approach to behavior learning for robots. In AT&P Journal Plus (2010-2), pp. 103-108. HMH s.r.o., Bratislava, Slovak Republik. ISSN 1336-5010. (Conference paper of RIE'10 which has also been selected by the conference committee to be published in AT&P Journal Plus magazine) [ bib ] M. Tokic. Adaptive ε-Greedy Exploration in Reinforcement Learning Based on Value Differences. In KI 2010: Advances in Artificial Intelligence, pp. 203-210. Springer Verlag, Heidelberg, Germany, 2010. [ bib ] [ pdf, the original publication is available at www.springerlink.com ] M. Tokic, A. Usadel, J. Fessler, and W. Ertel. On an educational approach to behavior learning for robots. In Proceedings of the 1st International Conference on Robotics in Education (RIE'10), pp. 171-176. Slovak University of Technology in Bratislava, Slovak Republik, 2010. ISBN 978-80-227-3353-3. [ bib ] [ pdf ] 2009 W. Ertel, M. Schneider, R. Cubek and M. Tokic. The Teaching-Box: A Universal Robot Learning Framework. In Proceedings of the 14th International Conference on Advanced Robotics (ICAR'09), pp. 1-6. IEEE Press, Munich, Germany, 2009. [ bib ] [ pdf, the original publication is available at ieeexplore.ieee.org] M. Tokic, W. Ertel, J. Fessler. The Crawler, A Class Room Demonstrator for Reinforcement Learning. In C.H. Lane & H.W. Guesgen (Eds.), Proceedings of the 22th International Florida Artificial Intelligence Research Society Conference (FLAIRS'09), pp. 160-165. AAAI Press, Menlo Park, California, USA, 2009. [ bib ] [ pdf, the original publication is available at aaai.org ] Errata: The interval of gamma should be [0,1) since the task is continuous (page 2+3). 2008 M. Tokic. Reinforcement learning an Robotern mit neuronalen Netzen. Masterthesis. University of Applied Sciences Ravensburg-Weingarten, Weingarten, Germany, 2008. [ bib ] 2007
	M. Tokic. Optimierung des Explorationsverhaltens eines lernenden Laufroboters. Scientific Project. University of Applied Sciences Ravensburg-Weingarten, Weingarten, Germany, 2007. [ bib ] 2006
	M. Tokic. Entwicklung eines lernenden Laufroboters, Diploma Thesis. University of Applied Sciences Ravensburg-Weingarten, Weingarten, Germany, 2006. [ bib ]
	M. Tokic, W. Ertel, H.-P. Radtke, J. Akmal, and W. Kroekel. Reinforcement learning on a simple real walking robot. In Proceedings of the 29th Annual German Conference on Artificial Intelligence, Bremen, Germany, 2006. [ bib ] [ pdf ]

Talks

M. Tokic. New Exploration Methods in Reinforcement Learning (invited talk), 1st International Workshop on Algorithmic Intelligence, Berlin Technical University, Germany, October 2011.
M. Tokic. Value-Difference Based Exploration: Adaptive Control between Epsilon-Greedy and Softmax, 34rd Annual German Conference on Artificial Intelligence (KI 2011), Berlin Technical University, Germany, October 2011.
M. Tokic. Reinforcement Lernen mit Wertunterschied-basierender Exploration, Ulm University, Germany, February 22nd 2011.
M. Tokic. On the importance of exploration/exploitation in reinforcement learning (invited lecture), Ravensburg-Weingarten University of Applied Sciences, Germany, Januar 2011.
M. Tokic. Adaptive ε-Greedy Exploration in Reinforcement Learning Based on Value Differences, 33rd Annual German Conference on Artificial Intelligence (KI 2010), Karlsruhe University, Germany, September 2010.
M. Tokic. On an educational approach for behavior learning for robots, 1st International Conference on Robotics in Education (RIE'2010), Bratislava University, Slovak Republik, September 2010.
M. Tokic. Einführung in das Q-Learning, Ravensburg-Weingarten University of Applied Sciences, Germany, November 2009.
M. Tokic. On VDBE Bandit Experiments and the Influence of alpha, Ulm University, Germany, October 2009.
M. Tokic. The Crawler, A Class Room Demonstrator for Reinforcement Learning, 22th International Florida Artificial Intelligence Research Society Conference (FLAIRS'2009), Sanibel Island, Florida, USA, May 2009.
[ pdf ]
M. Tokic. Exploration and Exploitation techniques in Reinforcement Learning (invited lecture), Ravensburg-Weingarten University of Applied Sciences, Germany, November 2008.
[ pdf ]
M. Tokic. REINFORCE-Algorithmen für das Erlernen von kontinuierlichen Aktionen, Ravensburg-Weingarten University of Applied Sciences, Germany, June 2008.
[ pdf ]
M. Tokic. Reinforcement learning on walking robots, Mannheim University of Applied Sciences, Germany, April 2008.