ABSTRACT
An optimum weight initialization which strongly improves the performance of the back propagation (BP) algorithm is suggested. By statistical analysis, the scale factor, R (which is proportional to the maximum magnitude of the weights), is obtained as a function of the paralyzed neuron percentage (PNP). Also, by computer simulation, the performances on the convergence speed have been related to PNP. An optimum range for R is shown to exist in order to minimize the time needed to reach the minimum of the cost function. Normalization factors are properly defined, which leads to a distribution of the activations independent of the neurons, and to a single nondimensional quantity, R, the value of which can be quickly found by computer simulation.
ABSTRACT
A digital computer simulation has been carried out of the exposure of a cell, modeled as a multilayered spherical structure, to an alternating electrical field. Electrical and electrochemical quantities of possible biological interest can be evaluated everywhere inside the cell. A strong frequency behavior in the range 0-10 MHz has been obtained.
Subject(s)
Cells/radiation effects , Electromagnetic Fields , Electromagnetic Phenomena , Models, Biological , Cell Membrane/radiation effects , Chromatin/radiation effects , Computers , Cytoplasm/radiation effects , Mathematics , Nuclear Envelope/radiation effectsABSTRACT
A digital computer simulation has been carried out of exposure of a cell, modelled as a multilayered spherical structure, to an alternating electrical field. Electrical quantities of possible biological interest can be evaluated everywhere inside the cell. A strong frequency selective behaviour in the range 0-10 MHz has been obtained.
Subject(s)
Biophysics , Electromagnetic Fields , Electromagnetic Phenomena , Biophysical Phenomena , Cell Membrane/physiology , Electric Conductivity , Ion Channels/physiology , Models, Biological , Nuclear Envelope/physiologyABSTRACT
A wide-band analysis is proposed about the behaviour of the dielectric constant of human sera, using a proper mathematical model. Each serum is identified by two parameters. There is a good agreement between theoretical and experimental data.
