Artificial Life and Robotics

, Volume 4, Issue 1, pp 42–45 | Cite as

Generating novel memories by integration of chaotic neural network modules

Original Article


A principle of integrating neural network modules based on chaotic dynamics was studied on our two-moduled Nozawa model. Chaotic neural networks represent each embedded pattern as a low-dimensional periodic orbit, and the others are shown as high-dimensional chaotic attractors. This is equivalent to W. Freeman’s “I don’t know” and “I know” states. In particular, we noted that the combination of two-way inputs to each neural network module conflicted with embedded Hebbian correspondence. It was found that the interaction between the modules generated a novel “I know” state in addition to the embedded representation. Chaotic neural network modules can autonomously generate novel memories or functions by this interaction. The result suggests a functional integration in neural networks as it ought to be, e.g., feature binding and gestalt.

Key words

Modularity Chaotic neural network Autonomous integration Hebbian learning 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Jacobs RA, Jordan MI (1991) Task decomposition through competition in a modular connectionist architecture: the what and where vision tasks. Cognitive Sci 15:219–250CrossRefGoogle Scholar
  2. 2.
    Kobayashi N, Iwata A (1993) Multi-module neural network model for higher order association. In: Proceedings IJCNN-93, vol I, pp 233–236Google Scholar
  3. 3.
    Happel BLM, Murre JMJ (1994) The design and evolution of modular neural network architectures. Neural Networks 7:985–1004CrossRefGoogle Scholar
  4. 4.
    Gallinari P (1995) Modular neural net systems: training of. In: Arbib MA (ed) The handbook of brain theory and neural networks. MIT Press, Cambridge, pp 582–583Google Scholar
  5. 5.
    Freeman WJ, Skarda CA (1985) Spatial EEG patterns, nonlinear dynamics and perception: the neo-Sherringtonian view. Brain Res Rev 10:147CrossRefGoogle Scholar
  6. 6.
    Tsuda I (1992) Dynamic link of memory—chaotic memory map in nonequilibrium neural networks. Neural Networks 5:313–326MathSciNetCrossRefGoogle Scholar
  7. 7.
    Adachi M, Aihara, K, Kotani M (1992) Nonlinear associative dynamics in a chaotic neural network. In: Proceedings of the 2nd International Conference on Fuzzy Logic and Neural Networks, pp 947–950Google Scholar
  8. 8.
    Nara S, Davis P, Totsuji H (1993) Memory search using complex dynamics in a recurrent neural network model. Neural Networks 6:963CrossRefGoogle Scholar
  9. 9.
    Freeman WJ (1994) Neural mechanisms underlying destabilization of cortex by sensory input. Physica D 75:151MATHCrossRefGoogle Scholar
  10. 10.
    Ishii S, Sato M (1998) Associative memory based on parametrically coupled chaotic elements. Physica D 121:344–366MATHMathSciNetCrossRefGoogle Scholar
  11. 11.
    Sano A (1998) Integrating neural network modules based on chaos, and its pattern recalling and internal dynamics (in Japanese). IEICE Technical Report, NC98-63Google Scholar
  12. 12.
    Sano A, Kunifuji S (1998) Integration of chaotic behavior modules into a flow-typed two-body neural network model. Electron Commun Jpn, Part III, 81(5):41–50CrossRefGoogle Scholar
  13. 13.
    Nozawa H (1992) A neural network model as a globally coupled map and applications based on chaos. Chaos 2:377–386MATHMathSciNetCrossRefGoogle Scholar
  14. 14.
    Hopfield JJ (1982) Neural networks and physical systems with emergent collective computational abilities. In: Proceedings of National Academy of Sciences USA, vol. 79, pp 2554–2558MathSciNetCrossRefGoogle Scholar
  15. 15.
    Aihara K, Takabe T, Toyoda M (1990) Chaotic neural networks. Phys Lett A 144:333MathSciNetCrossRefGoogle Scholar
  16. 16.
    Sano A (1995) The role of intersubsystem coupling density and chaotic dynamics in multi-body neural network model. In: Proceedings of IIW95, RWC Technical Report TR-95010, RWCP, pp 41–53Google Scholar
  17. 17.
    Hebb DO (1949) Organization of behavior. Wiley, New YorkGoogle Scholar

Copyright information

© ISAROB 2000

Authors and Affiliations

  1. 1.Graduate School of InformaticsKyoto UniversityKyotoJapan

Personalised recommendations