The Change of Memory Formation According to STDP in a Continuous-Time Neural Network Model

  • WATANABE Hidenori
    Department of Quantum Engineering and Systems Science, Faculty of Engineering, The University of Tokyo
  • WATANABE Masataka
    Department of Quantum Engineering and Systems Science, Faculty of Engineering, The University of Tokyo
  • AIHARA Kazuyuki
    Department of Mathematical Engineering and Information Physics, Faculty of Engineering, The University of Tokyo
  • KONDO Shunsuke
    Department of Quantum Engineering and Systems Science, Faculty of Engineering, The University of Tokyo

この論文をさがす

説明

Gerstner et al. proposed a learning rule, so called STDP (Spike-timing dependent synaptic plasticity). In this paper, Department of Quantum Engineering and Systems Science, Faculty of Engineering, The University of Tokyo we propose a continuous-time associative neural network model and study the function of STDP. Firstly, we show that our model is capable of acquiring new memory paterns by STDP. The memory patterns are retrieved as synchronous firing neurons. Secondly, we show that multiple patterns can be retrieved concurrently, which gives a solution to the superposition catastrophe. Furthermore, we show that an application of STDP to the concurrently retrieval of multiple patterns gives rise to the nested structure of memory.

収録刊行物

キーワード

詳細情報 詳細情報について

  • CRID
    1570572702512289792
  • NII論文ID
    110003219971
  • NII書誌ID
    AA10826272
  • ISSN
    09168532
  • 本文言語コード
    en
  • データソース種別
    • CiNii Articles

問題の指摘

ページトップへ