Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Model training method and device and service prediction method and device

A model training and model technology, applied in the computer field, can solve problems such as data heterogeneity is not considered, and the model cannot be better applied to local business

Pending Publication Date: 2022-06-28
ALIPAY (HANGZHOU) INFORMATION TECH CO LTD
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] However, at present, when modeling based on federated machine learning, the problem of data heterogeneity among various participants is not considered, resulting in the established model not being better applicable to the local business of the participants

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
  • Model training method and device and service prediction method and device
  • Model training method and device and service prediction method and device
  • Model training method and device and service prediction method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0047] As mentioned above, in the prior art, when modeling based on federated machine learning, the model used by each participant will eventually be updated to the converged global model generated by the server, that is, each participant will eventually The obtained model and the model used for actual business prediction are global models. In this way, the purpose of federated machine learning is achieved and the robustness of the model is improved.

[0048] However, there is a problem of data heterogeneity among the various participants, that is, the data structures of each participant are different, which makes the established model unable to better apply to the local business of the participants. Data heterogeneity can be reflected in the following five levels:

[0049] 1. Heterogeneity of computer architecture: The physical storage of data comes from computers with different architectures, such as mainframes, minicomputers, workstations, PCs or embedded systems.

[0050]...

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 provides a model training method and device based on federated machine learning and a service prediction method and device based on federated machine learning. The method comprises the following steps: training a public model by using local private data; uploading the public model to a server; receiving a global model issued by the server; the global model is aggregated by the server according to public models uploaded by at least two participants; updating a local public model by using the global model; and training the private model by using the local private data and the updated public model. The method and device provided by the embodiment of the invention can be better suitable for local services of participants.

Description

technical field [0001] One or more embodiments of this specification relate to computer technology, and in particular, to a private model training method and apparatus based on federated machine learning, and a business prediction method and apparatus. Background technique [0002] Federated machine learning is a distributed machine learning framework with privacy protection effect, which can effectively help multiple parties to conduct data usage and machine learning modeling while meeting the requirements of privacy protection, data security and government regulations. As a distributed machine learning paradigm, federated machine learning can effectively solve the problem of data islands, allowing participants to jointly model without sharing data, realize intelligent collaboration, and jointly train a global model with better performance. [0003] However, at present, when modeling based on federated machine learning, the problem of data heterogeneity among various partic...

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): G06N20/00
CPCG06N20/00Y02T10/40
Inventor 刘文鑫
Owner ALIPAY (HANGZHOU) INFORMATION TECH CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com