Home  News  Events  Album  Links  Japanese Version
Main Menu


Lost Password?

Register now!
2005-06-29: Human SCINT Seminar (7)
Poster Mihoko Otake  Registed 2005-12-25 22:26 (1652 hits)

Date: 2005.6.29 (Wed) 13:00-14:15
Place: General Research Building, Room 663
Speaker: Hideki Kadone
Title: Dynamical Systems Approach to Embodied Symbol Emergence
Keywords: Embodied Symbol Emergence, Dynamical Systems, Embodiment

Affiliation: Department of Mechano-Informatics,
Graduate School of Information Science and Technology
Position: Graduate Student
Adviser: Yoshihiko Nakamura, Nakamura Laboratory
Disciplines: Robotics, Neural Network, Dynamical Systems
Societies and Conferences: Robotics Society of Japan, Japan Neural Network Society

Bibliography: Hideki Kadone, Dynamical Systems Approach to Embodied Symbol Emergence, Human Science Integration Seminar Abstracts, No. 7, pp. 1, 2005.
(Please use this bibliography when you cite this abstract.)

Symbols are important for intelligent systems. Extracting important information from physical experiences and memorizing them as symbols enable one to apply the acquired information to other different situations. From this point of view, memory systems for humanoid robots are expected to have the function to self-organize symbols from accumulated associations of specific bodily memories.

We describe a memory system which forms symbolic representations and maintains specific-symbolic hierarchy based on stored memories of motion patterns, which was proposed by the presenter. The system consists of feature vectors of motions patterns based on auto-correlation of sequential postures, and an associative memory model with nonmonotonic activation function. The associative memory model encodes the symbols and the hierarchy by bifurcations attractors naturally mirroring the inherent hierarchical structure of the stored motion patterns.

In addition, as one of the potential physical bodily phenomena that make embodied symbol emergence possible, passive walking, which is a phenomenon that a compass-like mechanism walks down a slope without control with some stability, is discussed. Current situations of researches on them are reviewed.

[1] S.Harnad, "The Symbol Grounding Problem", Physica D, 42, pp.335-346, 1990.

[2] R.Brooks, "Intelligence without representation", Artificial Intelligence, 47, pp.139-159, 1991.

[3] Y.Kuniyoshi, M.Inaba, H.Inoue, "Leaning by Watching: Extracting Reusable Task Knowledge from Visual Obserbation of Human Performance", IEEE Trans. on Robotics and Automation, vol.10, no.6, pp.799-822,1994.

[4]T.Deacon, "Symbolic Species: The Co-Evolution of Language and the Brain", W.W.Norton and Company Inc., 1997.

[5] M.Donald, "Origins of the modern mind", Harvard University Press, 1991.

[6] T.Inamura, I.Toshima, H.Tanie, Y.Nakamura, "Embodied Symbol Emergence Based on Mimesis Theory", The International Journal of Robotics Research, Vol.23, No.4-5, pp.363-377, 2004.

[7] M.Okada, D.Nakamura, Y.Nakamura, "Self-organizing Symbol Acquisition and Motion Generation based on Dynamics-based Information Processing System", Proc. of the second International Workshop on Man-Machine Symbiotic Systems, pp.219-229, 2004.

[8] J.Tani, S.Nolfi, "Learning to perceive the world as articulated: an approach for hierarchical learning in sensory-motor systems", Neural Networks, 12, pp.1131-1141, 1999.

[9] M.Haruno, D.Wolpert, M.Kawato, "MOSAIC model for sensorimotor learning and control", Neural Computation, 13, pp.2201-2220, 2001.

[10] H.Kadone, Y.Nakamura, "Symbolic Memory for Humanoid Robots Using Hierarchical Bifurcations of Attractors in Nonmonotonic Neural Networks", submitted to IEEE/RSJ International Conference on Ingelligent Robots and Systems, 2005.

[11] S.Amari, "Neural Theory of Association and Concept-Formation", Biological Cybernetics, Vol.26, pp.175-185, 1977.

[12] E.Gardner, "Structure of metastable states in the Hopfield model", Journal of Physics A: Mathematical and General, L1047-L1052, 1986.

[13] T.Kimoto, M.Okada, "Mixed States on neural network with structural learning", Neural Networks, 17, 103-112, 2004.

[14] M.Morita, "Associative Memory with Nonmonotone Dynamics", Neural Networks, Vol.6, pp.115-126, 1993.

[15] M.Shiino, T.Fukai, "Self-consistent signal-to-noise analysis of the statistical behavior of analog neural networks and enhancement of the storage capacity", Physical Review E, 867-897, 1993.

[16] N.Matsumoto, M.Okada, Y.Sugase, S.Yamane, "Neuronal Mechanisms Encoding Global-to-Fine Infomation in Inferior-Temporal Cortex", Journal of Computational Neuroscience, 18, 85-103, 2005.

[17] T.McGeer, "Passive Dynamics Walking", International Journal of Robotics Research, Vol.9, pp.62-82, 1991.

[18] M.Garcia, A.Chatterjee, A.Ruina, M.Coleman, "The Simplest Walking Model: Stability, Complexity, and Scaling", ASME Journal of Biomechanical Engineering, Vol.120, No.2, pp.281-288, 1998.

[19] A.Goswami, B.Thuilot, B.Espiau, "A Study of the Passive Gait of a Compass-Like Biped Robot: Symmetry and Chaos", The International Journal of Robotics Research, pp.1282-1031, 1998.

[20] K.Osuka, K.Kirihara, "Motion Analysis and Experiments of Passive Walking Robot QUARTET II", Proc. of ICRA2000, pp.3052-3056, 2000.

[21] J.Hausdorff, C.Peng, Z.Ladin, J.Wei, A.Goldberger, "Is walking a random walk? Evidence for long-range correlations in stride interval of human gait", Journal of Applied Physiology, 78, pp.349-358, 1995.
Copyright (C) 2005-6, Human Science Integration Program - Humans. All right reserved.