Abstract
In this work we have implemented and tested a framework for training and analyzing auto-encoding architectures for the purpose of encoding state space representations into latent space, and thus compressing the essential space properties depicted in the training image dataset. We explored the possibility of incorporating a forward model in between encoder and decoder to predict the next state in series of temporally arranged images, with the aim of testing if it is able to predict meaningful representations (even if they are not interpretable by visual inspection). The application domain of this approach is that of cognitive robotics, where having a state representation learning system that can operate in an open-ended fashion is necessary. State representation learning has been a challenging topic in autonomous robotics for a long time, and some promising approaches have been developed. Anyway, when applying them to open-ended learning, as is the case of cognitive robotics, current solutions fail, so new systems must be implemented. The work presented here is a first approach towards this objective.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Achille, A., Soatto, S.: On the emergence of invariance and disentangling in deep representations. CoRR abs/1706.01350 (2017). http://arxiv.org/abs/1706.01350
Asada, M., et al.: Cognitive developmental robotics: a survey. IEEE Trans. Auton. Mental Dev. 1(1), 12–34 (2009). https://doi.org/10.1109/TAMD.2009.2021702
Goodfellow, I., Bengio, Y., Courville, A.: Deep Learning. MIT Press (2016). http://www.deeplearningbook.org
Ha, D., Schmidhuber, J.: World models. CoRR abs/1803.10122 (2018). http://arxiv.org/abs/1803.10122
Higgins, I., et al.: Towards a definition of disentangled representations. CoRR abs/1812.02230 (2018). http://arxiv.org/abs/1812.02230
Hjelm, R.D., et al.: Learning deep representations by mutual information estimation and maximization (2018)
Jonschkowski, R., Brock, O.: Learning state representations with robotic priors. Auton. Robots 39(3), 407–428 (2015). https://doi.org/10.1007/s10514-015-9459-7
Jonschkowski, R., Hafner, R., Scholz, J., Riedmiller, M.: PVEs: Position-velocity encoders for unsupervised learning of structured state representations (2017)
Kingma, D.P., Welling, M.: Auto-encoding variational bayes. CoRR abs/1312.6114 (2013). http://arxiv.org/abs/1312.6114
Kotseruba, I., Gonzalez, O.J.A., Tsotsos, J.K.: A review of 40 years of cognitive architecture research: focus on perception, attention, learning and applications. CoRR abs/1610.08602 (2016). http://arxiv.org/abs/1610.08602
Lesort, T., Rodríguez, N.D., Goudou, J., Filliat, D.: State representation learning for control: an overview. CoRR abs/1802.04181 (2018). http://arxiv.org/abs/1802.04181
Mnih, V., et al.: Human-level control through deep reinforcement learning. Nature 518(7540), 529–533 (2015). https://doi.org/10.1038/nature14236
Munk, J., Kober, J., Babuška, R.: Learning state representation for deep actor-critic control. In: 2016 IEEE 55th Conference on Decision and Control (CDC), pp. 4667–4673, December 2016. https://doi.org/10.1109/CDC.2016.7798980
Rumelhart, D.E., McClelland, J.L.: Learning Internal Representations by Error Propagation. MITP (1987). https://ieeexplore.ieee.org/document/6302929
Acknowledgements
This work has been partially funded by the Ministerio de Ciencia, Innovación y Universidades of Spain/FEDER (grant RTI2018-101114-B-I00) and Xunta de Galicia and FEDER (grant ED431C 2017/12).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Meden, B., Prieto, A., Peer, P., Bellas, F. (2020). First Steps Towards State Representation Learning for Cognitive Robotics. In: de la Cal, E.A., Villar Flecha, J.R., Quintián, H., Corchado, E. (eds) Hybrid Artificial Intelligent Systems. HAIS 2020. Lecture Notes in Computer Science(), vol 12344. Springer, Cham. https://doi.org/10.1007/978-3-030-61705-9_41
Download citation
DOI: https://doi.org/10.1007/978-3-030-61705-9_41
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-61704-2
Online ISBN: 978-3-030-61705-9
eBook Packages: Computer ScienceComputer Science (R0)