Abstract
Communication is not just the manipulation of words, but needs to decide what is communicated considering the surrounding situations and to understand the communicated signals considering how to reflect it on the actions. In this paper, aiming to the emergence of purposive and grounded communication, communication is seamlessly involved in the entire process consisted of one neural network, and no special learning for communication but reinforcement learning is used to train it. A real robot control task was done in which a transmitter agent generates two sounds from 1,785 camera image signals of the robot field, and a receiver agent controls the robot according to the received sounds. After learning, appropriate communication was established to lead the robot to the goal. It was found that, for the learning, the experience of controlling the robot by the transmitter is useful, and the correlation between the communication signals and robot motion is important.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Harnad, H.: Symbol Grounding Problem. Physica D 42, 335–346 (1990)
Nakano, K., Sakaguchi, Y., Isotani, R., Ohmori, T.: Self-Organizing System Obtaining Communication Ability. Biological Cybernetics 58, 417–425 (1988)
Steels, L.: Evolving grounded communication for robots. Trends in Cognitive Science 7(7), 308–312 (2003)
Werner, G.M.: DyerM.G.: Evolution of Communication in Artificial Organisms. In: Proc. of Artificial Life II, pp. 1–47 (1991)
Ono, N., et al.: Emergent Organization of Interspecies Communication in Q-Learning Artificial Organs. In: Advances in Artificial Life, pp. 396–405 (1995)
Shibata, K., Ito, K.: Emergence of Communication for Negotiation By a Recurrent Neural Network. In: Proc. of ISADS 1999, pp. 294–301 (1999)
Nakanishi, M., Shibata, K.: Effect of Action Selection on Emergence of One-way Communication Using Q-learning. In: Proc. of 10th AROB, CD-ROM, vol. GS7-3 (2005)
Shibata, K.: Discretization of Series of Communication Signals in Noisy Environment by Reinforcement Learning. Adaptive and Natural Computing Algorithms, 486–489 (2005)
Mitsunaga, N., et al.: Robovie-IV: A Communication Robot Interacting with People Daily in an Office. In: Proc. of IROS 2006, pp. 5066–5072 (2006)
Suga, Y., et al.: Development of Emotional Communication Robot, WAMOBEA-3. In: Proc. of ICAM 2004, pp. 413–418 (2004)
Bennewitz, M., et al.: Fritz - A Humanoid Communication Robot. In: Proc. of ROMAN 2007, pp. 1072–1077 (2007)
Shibata, K.: Emergence of Intelligence through Reinforcement Learning with a Neural Network. In: Advances in Reinforcement Learning, in Tech, pp. 99–120 (2011)
Barto, A.G., et al.: Neuronlike Adaptive Elements That can Solve Difficult Learning Control Problems’. IEEE Trans. of SMC 13, 835–846 (1983)
Rumelhart, D.E., et al.: Learning Internal Representation by Error Propagation. Parallel Distributed Processing (1986)
Nakanishi, M., et al.: Occurrence of State Confusion in the Learning of Communication Using Q-leaning. In: Proc. of 9th AROB, vol. 2, pp. 663–666 (2004)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2011 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Shibata, K., Sasahara, K. (2011). Emergence of Purposive and Grounded Communication through Reinforcement Learning. In: Lu, BL., Zhang, L., Kwok, J. (eds) Neural Information Processing. ICONIP 2011. Lecture Notes in Computer Science, vol 7064. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-24965-5_8
Download citation
DOI: https://doi.org/10.1007/978-3-642-24965-5_8
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-24964-8
Online ISBN: 978-3-642-24965-5
eBook Packages: Computer ScienceComputer Science (R0)