Abstract
This paper describes a softbot agent capable of learning to choose its actions, in order to achieve its goal when facing an opponent in a dynamic environment. The agent uses rewards gathered from the environment to assess and improve the quality of its own behavior. A multilayer perceptron neural network is assessed regarding its adequacy as a value function approximator for state-action pairs in the robotic soccer domain.
Leonardo Azevedo Scardua is supported by CNPq grant number 141802/97-9.
Anna H. Reali Costa is partially supported by FAPESP grant number 98/06417-9.
Jose Jaime da Cruz is partially supported by CNPq grant number 304071/85-4(RN).
Chapter PDF
Keywords
- Environment Reinforcement
- Learning Agent
- Reward Signal
- Average Goal
- Multilayer Perceptron Neural Network
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
References
Andre, D., Teller, A: Evolving Team Darwin United. Proceedings of the Second RoboCup Workshop, Paris 1998 317–323
Andre, D., Corten, E., Dorer, K., Gugenberger, P., Joldos, M., Kummeneje, J., Navratil, P., Itsuki, N., Riley, P., Stone, P., Takahashi, T., Yeap, T: Soccerserver Manual. 1999 http://www.dsv.su.se/johank/RoboCup/manual
Andou T: A RoboCup Team wich Reinforces Position Observationally. Proceedings of the Second RoboCup Workshop, Paris 1998 361–363
Bertsekas, D. and Tsitsiklis, John: Neuro Dynamic Programming Athena Scientific, Belmont, MA, 1996
Burkhar, H., Wendler, J., Gugenberger, P., Schroder, K., Kuhnel, R: AT-Humboldt in RoboCup-98. Proceedings of the Second RoboCup Workshop, Paris 1998 331–337
Haykin, S.: Neural Networks: a comprehensive foundation. 2nd ed., Prentice Hall, 1999
Kuzuaki, E., Sadaharu, I., Yamaguchi, H., Nobui, I. and Yoshiyuki, K: Team Description for Donguri. Proceedings of the Second RoboCup Workshop, Paris, 1998, 305–308
Matsumura T: Description of Team Erika. Proceedings of the Second RoboCup Workshop, Paris 1998 309–315
Hetch-Nielsen, R: Neurocomputing. Addison Wesley Publ. Co., New York, 1990
Sutton, R. and Barto, A: Reinforcement Learning: an introduction. MIT Press, 1998
Tesauro, G: Temporal Difference Learning and TD-Gammon. Communications of the ACM, 1995 Vol 38No. 3 58–68
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2000 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Scardua, L.A., Costa, A.H.R., da Cruz, J.J. (2000). Learning to Behave by Environment Reinforcement. In: Veloso, M., Pagello, E., Kitano, H. (eds) RoboCup-99: Robot Soccer World Cup III. RoboCup 1999. Lecture Notes in Computer Science(), vol 1856. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-45327-X_37
Download citation
DOI: https://doi.org/10.1007/3-540-45327-X_37
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-41043-0
Online ISBN: 978-3-540-45327-7
eBook Packages: Springer Book Archive