Simulation modeling of an analog impulse neural network based on a memristor crossbar using parallel computing technologies

Abstract

The work is devoted to the issues of simulation modeling of an analog impulse neural network based on memristive elements within the framework of the problem of pattern recognition. Simulation modeling allows you to configure the network at the level of a mathematical model, and subsequently use the obtained parameters directly in the process of operation. The network model is given as a dynamic system, which can consist of tens and hundreds of thousands of ordinary differential equations. Naturally, there is a need for an efficient and parallel implementation of an appropriate simulation model. OpenMP (Open Multi-Processing) is used as a technology for parallelizing calculations, since it allows you to easily create multi-threaded applications in various programming languages. The efficiency of parallelization is evaluated on the problem of modeling the process of learning the network to recognize a set of five images of size 128 by 128 pixels, which leads to the solution of about 80 thousand differential equations. On this problem, more than a sixfold acceleration of calculations was obtained.According to experimental data, the character of memristor operation is stochastic, as evidenced by the spread in the current-voltage characteristics during switching between high-resistance and low-resistance states. To take this feature into account, a memristor model with interval parameters is used, which gives upper and lower limits on the quantities of interest, and encloses the experimental curves in corridors. When modeling the operation of the entire analog self-learning impulse neural network, each epoch of training, the parameters of the memristors are set randomly from the selected intervals. This approach makes it possible to do without the use of a stochastic mathematical apparatus, thereby further reducing computational costs.