Abstract
Future robot systems will perform increasingly complex tasks in decreasingly well-structured and known environments. Robots will need to adapt their hardware and software, first only to foreseen, but ultimately to more complex changes of the environment. In this paper we describe a learning strategy based on reinforcement which allows fast robot learning from scratch using only its interaction with the environment, even when the reward is provided by a human observer and therefore is highly non-deterministic and noisy. To get this our proposal uses a novel representation of the action space together with an ensemble of learners able to forecast the time interval before a robot failure
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
Quintia, P., Iglesias, R., Rodriguez, M.A., Regueiro, C.V.: “Simultaneus learning of perception and action in mobile robots2. Robotics and Autonomous Systems 58(12), 1306–1315 (2010)
Lazebnik, S., Schmid, C., Ponce, J.: Beyond Bags of Features: Spatial Pyramid Matching for Recognizing Natural Scene Categories. In: Proceedings of the 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR 2006), pp. 2169–2178 (2006)
Kristo, K., Chua, C.S.: Image representation for object recognition: utilizing overlapping windows in Spatial Pyramid Matching. In: 20th IEEE International Conference on Image Processing (ICIP) (2013)
Quintia Vidal, P., Iglesias Rodriguez, R., Rodriguez Gonzalez, M.A., Vazquez Regueiro, C.: Learning on real robots from experience and simple user feedback. Journal of Physical Agents 7(1) (2013)
Carpenter, C.A., Grossberg, S., Rosen, D.B.: Fuzzy art: Fast stable learning and categorization of analog pattern by an adaptive resonance system. Neural Network 4(6), 759–771 (1991)
Barto, A.G.: Reinforcement learning: An introduction. MIT press (1998)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2015 Springer International Publishing Switzerland
About this paper
Cite this paper
Iglesias, R., Alvarez-Santos, V., Rodriguez, M.A., Santos-Saavedra, D., Regueiro, C.V., Pardo, X.M. (2015). Pyramid Representations of the Set of Actions in Reinforcement Learning. In: Ferrández Vicente, J., Álvarez-Sánchez, J., de la Paz López, F., Toledo-Moreo, F., Adeli, H. (eds) Bioinspired Computation in Artificial Systems. IWINAC 2015. Lecture Notes in Computer Science(), vol 9108. Springer, Cham. https://doi.org/10.1007/978-3-319-18833-1_22
Download citation
DOI: https://doi.org/10.1007/978-3-319-18833-1_22
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-18832-4
Online ISBN: 978-3-319-18833-1
eBook Packages: Computer ScienceComputer Science (R0)