A Study on High-precision Analog Neural Network VLSI Computers

Bibliographic Information

Other Title
  • アナログニューロVLSIコンピュータの高精度化に関する検討 : 対比誤差逆伝播学習によるオンチップ学習の高効率化

Search this article

Description

This paper demonstrates usefulness of analog neuro-LSIs to construct general neural networks with analog dynamics.From a perspective of calculation resolution,analog LSI implementation is more suitable for high speed neuraj networks than digital LSI implementation.It is also demonstrated that noise in analog circuits less affects the learning performance than weight quantization in digital memories.It is described that serious effects of offset errors in analog circuits on the backpropagation(BP)learning performance can be lowered by using a new learning algorithm called contrastive BP learning,and a circuit architecture implementing the learning procedure is explained.

Journal

  • Technical report of IEICE. ICD

    Technical report of IEICE. ICD 93 (231), 23-30, 1993-09-17

    The Institute of Electronics, Information and Communication Engineers

Keywords

Details 詳細情報について

  • CRID
    1571980077366395776
  • NII Article ID
    110003316880
  • NII Book ID
    AN10013276
  • Text Lang
    ja
  • Data Source
    • CiNii Articles

Report a problem

Back to top