Multi-mode neural morphological network core

Abstract

Provided in the invention is a multi-mode neural morphological network core comprising a mode register, an axon input unit, a synaptic weight storage unit, a dendrite unit and a neuron computing unit. According to the multi-mode neural morphological network core, both artificial neural network computing and impulsive neural network computing can be carried out; and switching between an artificial neural network operation mode and an impulsive neural network operation mode can be realized according to demands.

Description

technical field [0001] The present invention relates to a neuromorphic network kernel. Background technique [0002] A neural network is a computing system that imitates the synapse-neuron structure of a biological brain for data processing. It consists of multi-layered computing nodes and connections between layers. Each node simulates a neuron and performs a specific operation, such as an activation function. The connection between nodes simulates a synapse, and the weighted value of the connection to the input from the upper layer node represents the synaptic weight. Neural networks have powerful nonlinear and adaptive information processing capabilities. [0003] The neuron in the artificial neural network processes the accumulated value from the connection input with the activation function as its own output. Corresponding to different network topologies, neuron models and learning rules, artificial neural networks include dozens of network models such as perceptrons,...

AI Technical Summary

Problems solved by technology

However, the discontinuity of the neuron model of the spiking neural network, the complexity of space-time coding, and the uncertainty of the network structure make it difficult to describe the overall network mathematically, so it is difficult to construct an effective and general supervised learning algorithm. Limits its computational scale and accuracy

Images