A parameter tuning method for PQN model

DOI
  • Sakai Daimon
    Department of Information Science and Technology, The University of Tokyo
  • Nanami Takuya
    Institute of Industrial Science, The University of Tokyo
  • Kohno Takashi
    Institute of Industrial Science, The University of Tokyo

Description

The Piecewise Quadratic Neuron (PQN) model is a spiking neuron model that can be efficiently implemented on digital arithmetic circuits. In addition, this model can reproduce a variety if neuronal activities precisely with optimized parameter sets. In previous studies, we have optimized the parameters using meta-heuristic methods, which required a lot of computational time. In this paper, we proposed an parameter fitting method that takes into account the mathematical structure of the model and reproduces the electrophysiological activities of a target neuron with less computational time. We expect that this method can be used to construct silicon neuronal networks that faithfully replicate the nervous system. This method is expected to applicable to building silicon neuronal networks that faithfully replicate the nervous system.

Journal

Keywords

Details 詳細情報について

  • CRID
    1390291767555777280
  • DOI
    10.5954/icarob.2022.os15-2
  • ISSN
    21887829
  • Text Lang
    en
  • Data Source
    • JaLC
    • Crossref
    • OpenAIRE
  • Abstract License Flag
    Disallowed

Report a problem

Back to top