Unlock instant, AI-driven research and patent intelligence for your innovation.

Federated learning-based resource allocation optimization method and system

A technology of resource allocation and optimization method, applied in the field of machine learning, it can solve problems such as large time difference, non-IID distribution, and model non-convergence, and achieve the effect of improving speed and reducing delay.

Active Publication Date: 2021-08-27
PEKING UNIV
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the data set of the user equipment has the characteristics of non-independent and identical distribution, which will make the local models of different user equipment and some models of the edge server reach the target accuracy. The time is very different, and even some models do not converge.

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
  • Federated learning-based resource allocation optimization method and system
  • Federated learning-based resource allocation optimization method and system
  • Federated learning-based resource allocation optimization method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0021] Such as figure 1 As shown, this embodiment provides a method for optimizing resource allocation based on federated learning, and the method includes:

[0022] Step 101: Randomly assign each user equipment to an edge server;

[0023] Step 102: Independently and identically distributed adjustments are made to the connection modes between the user equipment and the edge server to obtain an optimal resource allocation mode.

[0024] When adjusting the independent and identical distribution of the connection mode between the user equipment and the edge server, it needs to be adjusted according to the optimization function. Therefore, when allocating user devices to edge servers, the first thing to do is to build an optimization function. Specifically, in this embodiment, the system delay is used as the optimization target, and the optimization variables are the processor duration t of the user equipment and the channel bandwidth allocation coefficient b, then the optimizat...

Embodiment 2

[0048] Such as image 3 As shown, this embodiment provides a resource allocation optimization system based on federated learning, and the system includes:

[0049] The random assignment module M1 is used to randomly assign each user equipment to an edge server;

[0050] The optimization module M2 is configured to perform independent and same-distributed adjustment on the connection mode between the user equipment and the edge server to obtain an optimal resource allocation mode.

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 invention relates to a federated learning-based resource allocation optimization method and system. The method comprises the steps of carrying out the adjustment of a connection relation between user equipment and edge servers, thereby enabling all data sets of the user equipment covered by each edge server to be close to independent identical distribution, enabling the speed of each part model to reach the target precision to be high. Therefore, the delay can be reduced to the greatest extent. According to the invention, better system resource allocation is realized, and lower system delay is realized at the same time.

Description

technical field [0001] The present invention relates to the technical field of machine learning, in particular to a resource allocation optimization method and system based on federated learning. Background technique [0002] Applying federated learning technology to three-tier edge computing can avoid direct upload of private data of user equipment, thereby protecting user data privacy. Since the edge server has lower latency than the cloud server, in the three-tier system, some models can be integrated through the edge server to achieve higher efficiency. But in a three-tier system, the computing and communication resources of edge servers and user equipment are limited, and better algorithms need to be designed to achieve optimal resource allocation to achieve the lowest system delay. However, the data set of the user equipment has the characteristics of non-independent and identical distribution, which will make the local models of different user equipment and some mode...

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 & Authority Applications(China)
IPC IPC(8): G06F9/50G06N20/20
CPCG06F9/5027G06F9/5072G06N20/20G06F2209/502
Inventor 宋令阳刘天宇安鹏边凯归程翔孙绍辉庹虎
Owner PEKING UNIV