Self-adaptive threshold value neuron information processing method and system

Abstract

The invention relates to a self-adaptive threshold value neuron information processing method and system. The method comprises the steps that the front-end pulse neuron output information is received; the current pulse neuron information is read; the current pulse neuron output information is calculated based on the front-end pulse neuron output information and the current pulse neuron information; the current self-adaptive threshold value variable and threshold value potential are read, and the current self-adaptive threshold value is calculated; if the current pulse neuron output information is greater than or equal to the self-adaptive threshold value, the current pulse neuron output information is output, and the current self-adaptive threshold value variable is updated according to a first self-adaptive threshold value updating model; and otherwise the current pulse neuron output information is not output, and the current self-adaptive threshold value variable is updated according to a second self-adaptive threshold value updating model. According to the invention, the grant frequency of each neuron in the whole network can be effectively balanced, and the information processing capability of the pulse neural network is improved.

Description

technical field [0001] The invention relates to the technical field of artificial neural networks, in particular to an adaptive threshold neuron information processing method and system. Background technique [0002] Most of today's artificial neural network research is still implemented in von Neumann computer software and high-performance GPGPU (General Purpose Graphic Processing Units) platform. The hardware overhead, energy consumption and information of the whole process The processing speed is not optimistic. For this reason, the field of neuromorphic computing has developed rapidly in recent years, that is, using hardware circuits to directly construct neural networks to simulate the functions of the brain, trying to achieve a computing platform that is massively parallel, low-energy, and capable of supporting complex pattern learning. [0003] However, in traditional neuromorphic systems, how to effectively balance the firing frequency of each spiking neuron in the ...

AI Technical Summary

Problems solved by technology

[0004] Based on this, it is necessary to provide an adaptive threshold neuron information processing method and system for how to effectively balance the firing frequency of each spike neuron in the entire neural network. The method includes:

Images