Edge computing node distribution and quit method based on branch neural network

Abstract

The invention discloses an edge computing node distribution and quit method based on a branch neural network, which is mainly divided into three steps in order to improve the safety of an artificial intelligence model under edge computing and accelerate the computing efficiency of the model: training a neural network model, deploying a branch neural network model at an edge computing node and selecting a model exit point: firstly building the neural network model and combining with a model rectification algorithm to train the model; secondly, in the process of distributing edge computing nodes for the trained classification model, selecting proper edge computing nodes by utilizing a minimum delay algorithm; and finally, selecting an appropriate model exit point in a model inference stage to reduce the calculation amount of an edge calculation node, so that the precision and accuracy of classification network prediction are improved, the effects of improving the safety of a neural network model and accelerating the calculation efficiency of the model are finally achieved, and the purposes of defending against samples and improving the calculation efficiency of the model are achieved.

Description

technical field [0001] The present invention relates to the technical field of a computer system based on a specific computing model, specifically a method for assigning and withdrawing edge computing nodes based on a branched neural network, which is applied to strengthen the security of the neural network model under the framework of edge computing and confront Defense of image samples and improving computational efficiency of models. Background technique [0002] The edge computing model is an emerging technology whose vulnerabilities have not been fully explored. The few studies on edge computing attacks mainly focus on possible threats to sensor networks, including some mobile phones and PCs. [0003] A security issue arising under edge computing, insufficient validation of input data may lead to malicious injection attacks. This malicious injection of data is a relatively common attack method for edge computing. The attacker can inject malicious input to cause the se...

AI Technical Summary

Problems solved by technology

However, due to the limited processing power of current edge computing devices, a single edge computing node may not be able to complete the inference task of complex network models, so multiple edge computing nodes are required to jointly deploy the DNN model

Images