Flexible memristor based neuromorphic system for implementing multi-layer neural network algorithms

Abstract

This paper describes a memristor-based neuromorphic system that can be used for ex situ training of various multi-layer neural network algorithms. This system is based on an analogue neuron circuit that is capable of performing an accurate dot product calculation. The presented ex situ programming technique can be used to map many key neural algorithms directly onto the grid of resistances in a memristor crossbar. Using this weight-to-crossbar mapping approach along with the memristor based circuit architecture, complex neural algorithms can be easily implemented using this system. Some existing memristor based circuits provide an approximated dot product based on conductance summation, but neuron outputs are not directly correlated to the numerical values obtained in a traditional software approach. To show the effectiveness and versatility of this circuit, two different powerful neural networks were simulated. These include a Restricted Boltzmann Machine for character recognition and a Multilayer Perceptron trained to perform Sobel edge detection. Following these simulations, an analysis was presented that shows how both memristor accuracy and neuron circuit gain relates to output error.

This work presents a novel memristor based architecture that that is capable of implementing multiple different learning algorithms using the same hardware, which is based on crossbar structures such as the one displayed. The example presented shows the result of the memristor architecture when implementing Sobel edge detection using a multilayer perceptron.

Keywords:

• Memristor
• neural network
• neuromorphic
• hardware
• dot product