Special-purpose brainware architecture for data processing

  • Tadashi Ae
  • Hikaru Fukumoto
  • Saku Hiwatashi
Innovative Architectures
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1259)


A new architecture SBA is proposed for real-time data processing. The SBA is derived from systematic realization of human brain, two important features (adaptiveness and stability) of which are carefully introduced. The scheme of dataflow on SBA seems to be similar to that of a conventional parallel machine, but the architecture is different in realizing directly two types of memory, i.e., STM (Short Term Memory) and LTM (Long Term Memory). Learning is performed on these two memories, since SBA supports both types of learning; one for STM and the other for LTM.

As a result, the SBA may become a new type of neural AI (Artificial Intelligence) computer.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    M. Ito: Expecting for Brain Research in the 21st Century, Extended Abstract of Brainware Workshop, Tsukuba, FED-148, pp. 3–5 (1996, in Japanese).Google Scholar
  2. 2.
    G. Matsumoto, Y. Shigematsu, M. Ichikawa: Brain-computing, i.b.i.d. pp. 31–36 (1996, in Japanese).Google Scholar
  3. 3.
    R. Agrawal: Database Mining: A Performance Perspective, IEEE Transaction on Knowledge and Data Engineering, Vol. 5, No. 6, pp. 914–925 (1993).Google Scholar
  4. 4.
    H. Araki, H. Fukumoto, T. Ae: Image Processing using Simplified Kohonen Network, Proceedings SPIE, San Jose, Vol. 2661, pp. 24–33 (1996).Google Scholar
  5. 5.
    J. J. Hopfield: Neurons with Graded Response Have Collective Computational Properties like Those of Two-State Neurons, Proceedings National Academy of Science, USA81, pp. 3088–3092 (1984).Google Scholar
  6. 6.
    K. Aihara: Chaos and Brainware, Extended Abstract of Brainware Workshop, Tsukuba, FED-148, pp. 8–12 (1996).Google Scholar
  7. 7.
    K. Murota, X. Wu, T. Ae: High-Speed Processing for Knowledge Representation from Decision Diagram, JSAI Technical Report, SIG-PPAI-9502-2 (1995, in Japanese).Google Scholar
  8. 8.
    H. Fukumoto, S. Hiwatashi, X. Wu, T. Ae: Knowledge Discovery by Neural Reasoning, JSAI Technical Report, SIG-PPAI-9503-6 (1996, in Japanese).Google Scholar
  9. 9.
    T. Ae, T. Toyosaki, H. Fukumoto, K. Sakai: ONBAM: An Objective-Neuron-Based Active Memory, Proceedings 1st ICA3PP, Brisbane, Vol. 1, pp. 231–234 (1995).Google Scholar
  10. 10.
    T. Toyosaki, T. Ae: A Neuro-Chip for Kohonen's LVQ Algorithm, Proceedings 6th ISIC, Singapore, pp. 107–111 (1995).Google Scholar
  11. 11.
    T. Ae: New Functional Machine, (New) Information Processing Handbook (JIPS Ed.), 3.8.7., pp. 455–459, Ohm Co. Ltd. (1995, in Japanese).Google Scholar
  12. 12.
    M. Kawada, X. Wu, T. Ae: A Construction of Neural-Net Based AI Systems, Proceedings 1st ICECCS, Ft. Lauderdale, pp. 424–427 (1995).Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1997

Authors and Affiliations

  • Tadashi Ae
    • 1
  • Hikaru Fukumoto
    • 1
  • Saku Hiwatashi
    • 1
  1. 1.Electrical Engineering, Faculty of EngineeringHiroshima UniversityHigashi-HiroshimaJapan

Personalised recommendations