Title :
Temporal resolution: a critical parameter in simulations of pulse-coupled neural networks
Author :
Kraut, Ulrike ; Nischwitz, Alfred ; Waschulzik, Thomas
Author_Institution :
Lehrstuhl fur Nachrichtentech., Tech. Univ. Munchen, Germany
Abstract :
The synchronization of impulse generation in non-oscillating networks of laterally coupled model neurons was subject of former investigations. The coupled nonlinear equation system describing the dynamics of the neural network was approximated by the Euler-method. In this work, the influence of the size of the increment of the Euler-approximation on synchronization results is investigated. It turns out that the deviations in the synchronization quality for different increments depend on the network parameters and all results taken from temporally discrete systems must be treated with extreme caution.
Keywords :
approximation theory; discrete systems; dynamics; neural nets; neurophysiology; physiological models; synchronisation; Euler-approximation; critical parameter; impulse generation; network dynamics; neuron model; pulse-coupled neural networks; simulations; synchronization; temporally discrete systems; Couplings; Differential equations; Electronic mail; Electrophysiology; Intelligent networks; Neural networks; Neurons; Nonlinear dynamical systems; Nonlinear equations; Pulse generation;
Conference_Titel :
Neural Networks, 1993. IJCNN '93-Nagoya. Proceedings of 1993 International Joint Conference on
Print_ISBN :
0-7803-1421-2
DOI :
10.1109/IJCNN.1993.716723