A spiking neural network model with fuzzy learning rate application for complex handwriting movements generation

Abstract

In this paper a spiking neural network model with fuzzy learning rate for online complex handwriting movement generation is proposed. The network is composed of an input layer which uses a set of Beta-elliptic parameters as input, a hidden layer and an output layer dealing with the estimation of the script coordinates X(t) and Y (t). An additional input is used as a timing network to prepare the input parameters. We also propose a Fuzzy Learning Rate (FLR) for our spiking neural network. This rate is obtained by combining an Adaptive Learning Rate (ALR) with a fuzzy logic based supervisor. The obtained results showed the efficiency of the proposed fuzzy strategy for the online adjustment of the learning rate. Indeed, we have improved, indifferently from the initialization, the Neural Network training quality in terms of rapidity and precision. Similarity degree is measured between original and generated scripts to evaluate our model.