Integrity Intelligent Network Training Method Based on Edge Collaboration

An intelligent network and training method technology, applied in neural learning methods, biological neural network models, instruments, etc., can solve problems such as waste of resources, high requirements for computing resources, edge nodes reducing the overall performance of edge intelligent networks, etc., to reduce the impact , the effect of improving training efficiency

Active Publication Date: 2022-03-29
派欧云计算(上海)有限公司
View PDF12 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Edge intelligent networks generally use distributed methods to complete training through several edge nodes. However, existing distributed methods usually train the same model, use different training data sets, and execute in different fields, without considering the separability of deep networks. ; At the same time, due to the security problems of the edge nodes, such as data loss in the communication process of the edge nodes, etc., the security problems of these edge nodes will have a negative impact on the overall performance of the edge intelligent network
[0004] In the process of realizing the present disclosure, it is found that when training a deep network in the prior art, the computing resources of edge nodes performing training are relatively high, and the application of edge nodes with low computing resources is lacking, resulting in a large waste of resources; At the same time, the existing technology ignores the investigation of the reliability of edge nodes, resulting in the addition of new edge nodes that may degrade the overall performance of the edge intelligent network

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Integrity Intelligent Network Training Method Based on Edge Collaboration
  • Integrity Intelligent Network Training Method Based on Edge Collaboration
  • Integrity Intelligent Network Training Method Based on Edge Collaboration

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0024] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. It should be understood, however, that these descriptions are exemplary only, and are not intended to limit the scope of the present disclosure. In the following detailed description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the embodiments of the present disclosure. It may be evident, however, that one or more embodiments may be practiced without these specific details. Also, in the following description, descriptions of well-known structures and techniques are omitted to avoid unnecessarily obscuring the concept of the present disclosure.

[0025] The terminology used herein is for the purpose of describing particular embodiments only, and is not intended to be limiting of the present disclosure. The terms "comprising", "comprising", etc. used herein indicate the presence of stated features, ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The present disclosure provides a method for training an honest intelligent network based on edge collaboration, including: dividing a deep network model into multiple sub-models, and assigning the multiple sub-models to multiple edge nodes; The sub-model trained in the node; the adjusted sub-model is trained, and the trained sub-model is aggregated in the task node to obtain the updated deep network model; and the loss value of the updated deep network model is adjusted based on the edge node. Reputation value, and finally get the integrity intelligent network.

Description

technical field [0001] The present disclosure relates to the field of intelligent network technology / edge computing technology, and more specifically, to a method for training an honest intelligent network based on edge collaboration. Background technique [0002] In the past few years, AI (Artificial Intelligence, artificial intelligence) technology has made breakthrough progress. Artificial intelligence can improve the automation of machines and improve the utilization of resources, which brings great potential for our economic development. Edge intelligent networks can meet user demands for high data rates, ubiquitous accessibility, and low response latency. [0003] Edge intelligent networks generally use distributed methods to complete training through several edge nodes. However, existing distributed methods usually train the same model, use different training data sets, and execute in different fields, without considering the separability of deep networks. ; At the ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & AuthorityPatents(China)
IPC IPC(8): G06K9/62G06N3/04G06N3/08
CPCG06N3/084G06N3/045G06F18/214
Inventor齐昊天仇超刘志成王晓飞
Owner派欧云计算(上海)有限公司