Parameter estimation of various Hodgkin-Huxley-type neuronal models using a gradient-descent learning method

摘要

The automatic parameter identification method proposed by Doya et al. (1994) of the Hodgkin-Huxley-type equations (1952) is investigated in detail. The Hodgkin-Huxley-type equations describe membrane currents and conduction and excitation in nerves. An improved estimation method is proposed and it is shown that our method resolves the difficulties in estimating parameters of such equations with complicated membrane potential waveforms such as a chaotic bursting and also much improves the parameter estimation (learning) speed.