Abstract
We present a model of Reinforcement Learning, which consists of modified neural-network architecture with spatio-temporal connections, known as Temporal Hebbian Self-Organizing Map (THSOM). A number of experiments were conducted to test the model on the maze solving problem. The algorithm demonstrates sustainable learning, building a near to optimal routes. This work describes an agents behavior in the mazes of different complexity and also influence of models parameters at the length of formed paths.
References
Mnih, V., Kavukcuoglu, K., Silver, D., Graves, A., Antonoglou, I., Wierstra, D., Riedmiller, M.: Playing Atari with Deep Reinforcement Learning (2013)
Koutník, J., Šnorek, M.: Temporal Hebbian Self-Organizing Map for Sequences (2008)
Gupta, S., et al.: Cognitive Mapping and Planning for Visual Navigation. arXiv:1702.03920
Schrodt, F., et al.: Mario Becomes Cognitive. Top. Cogn. Sci. (2017). p. 131
Paxton, C., et al.: Combining Neural Networks and Tree Search for Task and Motion Planning in Challenging Environments (2017). arXiv:1703.07887
Broy, M.: Software engineering – from auxiliary to key technologies. In: Broy, M., Dener, E. (eds.) Software Pioneers, pp. 10–13. Springer, Heidelberg (2002)
Panov, A.I.: Behavior planning of intelligent agent with sign world model. Biol. Inspired Cogn. Archit. 19, 21–31 (2017)
Emelyanov, S., et al.: Multilayer cognitive architecture for UAV control. Cogn. Syst. Res. 39, 58–72 (2016)
Kaelbling, L.P., Littman, M.L., Moore, A.W.: Reinforcement learning: a survey. J. Artif. Intell. Res. 4, 237–285 (1996)
Chalita, M.A., Lis, D., Caverzasi, A.: Reinforcement learning in a bio-connectionist model based in the thalamo-cortical neural circuit. Biol. Inspired Cogn. Archit. 16, 45–63 (2016)
Acknowledgements
The reported study was supported by RFBR, research Projects No. 16-37-60055 and No. 15-07-06214.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer International Publishing AG
About this paper
Cite this paper
Filin, D., Panov, A.I. (2018). Applying a Neural Network Architecture with Spatio-Temporal Connections to the Maze Exploration. In: Samsonovich, A., Klimov, V. (eds) Biologically Inspired Cognitive Architectures (BICA) for Young Scientists. BICA 2017. Advances in Intelligent Systems and Computing, vol 636. Springer, Cham. https://doi.org/10.1007/978-3-319-63940-6_8
Download citation
DOI: https://doi.org/10.1007/978-3-319-63940-6_8
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-63939-0
Online ISBN: 978-3-319-63940-6
eBook Packages: EngineeringEngineering (R0)