Federal learning method, device and equipment

A learning method and federated technology, applied in the field of machine learning, can solve problems such as inability to exchange data with each other, achieve the effect of efficient learning process, reduce cloud pressure, and reduce the number of

Active Publication Date: 2021-06-25
ANT FINANCIAL HANG ZHOU NETWORK TECH CO LTD
View PDF9 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, due to privacy constraints and trust issues among various institutions, data cannot be exchanged with each other, forming a data island

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
  • Federal learning method, device and equipment
  • Federal learning method, device and equipment
  • Federal learning method, device and equipment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0065] Embodiments of this specification provide a federated learning method, device, device, and storage medium.

[0066] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described The embodiments are only some of the embodiments of the present application, but not all of them. Based on the embodiments of this specification, all other embodiments obtained by persons of ordinary skill in the art without creative efforts shall fall within the scope of protection of this application.

[0067] In one or more embodiments of this specification, clustering is performed in advance according to the similarity between different clients to obtain corresponding client groups, and a representative client that can represent the c...

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 embodiment of the invention discloses a federal learning method, device and equipment. According to the scheme, the method comprises the following steps: determining a plurality of client groups obtained by clustering according to the similarity between clients and representative clients in the client groups; sending a training request to the client group by the representative client; receiving training data of a part of clients in the client group fed back by the representative client, wherein the part of clients are determined by a pre-trained elastic uploading prediction model according to client states; and determining gradient data through the training data, and sending the gradient data to the representative client, so that the clients in the client group perform gradient updating.

Description

technical field [0001] This description relates to the technical field of machine learning, and in particular to federated learning methods, devices and equipment. Background technique [0002] With the development of computer technology, machine learning is playing an increasingly important role in all walks of life, and the performance of machine learning models also increases with the increase in the amount of data. However, due to privacy constraints and trust issues among various institutions, data cannot be exchanged with each other, forming a data island. To solve this problem, the concept of federated learning is proposed. That is, using a trusted third party, each institution transmits the data to the third party, and the third party calculates the gradient in the cloud, and then sends it to each institution, and each institution can update the model. In this way, various institutions get gradients calculated from more data without direct data exchange, thereby im...

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): G06N20/00G06N3/04G06N3/08G06K9/62G16Y20/00G16Y20/10G16Y20/40G16Y40/10G16Y40/20
CPCG06N20/00G06N3/08G16Y20/00G16Y20/10G16Y20/40G16Y40/10G16Y40/20G06N3/045G06F18/231G06F18/22G06F18/23G06F18/23213G06F18/2411G06F18/24Y02D10/00
Inventor 曹佳炯丁菁汀
Owner ANT FINANCIAL HANG ZHOU NETWORK TECH CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap