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

Federated learning system

A learning system and federated technology, applied in the field of artificial intelligence, can solve the problems of low efficiency of federated learning and large amount of data, etc.

Pending Publication Date: 2021-12-07
BEIJING WODONG TIANJUN INFORMATION TECH CO LTD +1
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In the process of realizing the present invention, the inventor found that there are at least the following technical problems in the prior art: in the prior art, the amount of data to be transmitted during each iteration of the modeling process is relatively large, and the efficiency of federated learning is low

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 system
  • Federated learning system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0034] figure 1 It is a training flowchart of a federated learning system provided by the embodiment of the present invention. This embodiment is applicable to the situation of training a federated learning system, especially applicable to the situation of training a vertical federated learning system.

[0035] In this embodiment, the federated learning system includes a coordinating terminal and at least two participating terminals. The participating terminal includes a label holding terminal and at least one feature holding terminal. The coordinating terminal and each participating terminal are connected by communication. Taking the risk control scenario as an example, its characteristic is that the tag is only held by one of the participants (which can be called Guest), while the other participants only have some characteristics of the data (which can be called Host). Through the cooperation of Guest and Host, the effect of the model can be improved, and the purpose of red...

Embodiment 2

[0064] This embodiment provides a preferred embodiment on the basis of the foregoing embodiments. In this embodiment, the tag holder is embodied as a Guest, the feature holder is embodied as a Host, and the collaboration end is embodied as a third-party Coordinator.

[0065] In this embodiment, it is assumed that there are P participating terminals, one of which is a Guest, and the remaining p-1 participating terminals are Hosts, and there is also a third-party Coordinator to coordinate the entire process. Note that the local data of the pth participant after private data alignment is x p , the model parameter is w p . Before iterative training, each participant needs to be initialized. Specifically, each participant generates an initial model parameter w p (0), and then calculate the intermediate value locally, and each Host transmits the ciphertext to the Guest through additive homomorphic encryption (such as Paillier). After the Guest gets all the intermediate values, ...

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 federated learning system, which comprises a cooperation terminal and at least two participation terminals, each participation terminal comprises a label holding terminal and at least one feature holding terminal, the cooperation terminal is in communication connection with each participation terminal, and training of the federated learning system comprises the following steps: determining one participation terminal as a current model updating terminal; obtaining model updating parameters by the current model updating terminal, determining an encryption gradient value according to the model updating parameters and sending the encryption gradient value to the cooperation terminal, and determining the model updating parameters by the label holding end according to local model parameters of the previous current model updating terminal; determining a model updating direction by the cooperation terminal according to the encryption gradient value and sending the model updating direction to the current model updating terminal; and updating the local model parameters by the current model updating terminal according to the model updating direction. By updating the local models of the participants one by one, the participants are prevented from transmitting a large amount of data to the cooperation terminal at the same time, and the federal learning efficiency is improved.

Description

technical field [0001] Embodiments of the present invention relate to the technical field of artificial intelligence, and in particular to a federated learning system. Background technique [0002] Federated learning refers to a machine learning framework that can effectively help multiple participants (which can represent individuals or institutions) jointly train models while meeting the requirements of data privacy protection. According to the different relationships between samples and features of data held by each participating end of federated learning, it can be divided into horizontal federated learning, vertical federated learning, and federated transfer learning. In some scenarios, vertical federated learning has been more widely used. Vertical federated learning modeling requires encrypted alignment of samples first, and then encrypted model training. During the modeling process, each participating end sends information to a third-party collaboration end, that i...

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): G06N20/20G06K9/62
CPCG06N20/20G06F18/214
Inventor 杨恺王虎黄志翔彭南博
Owner BEIJING WODONG TIANJUN 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