Model training method and device for protecting privacy in distributed system

A distributed system, privacy protection technology, applied in the field of machine learning, to improve the efficiency of data transmission, protection from leakage, and good prediction performance

Active Publication Date: 2020-12-18
ALIPAY (HANGZHOU) INFORMATION TECH CO LTD
View PDF4 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Therefore, in the scenario where multiple parties jointly train the model, data security and data privacy issues are a great challenge

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 for protecting privacy in distributed system
  • Model training method and device for protecting privacy in distributed system
  • Model training method and device for protecting privacy in distributed system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment approach

[0096]According to an embodiment, it is assumed that the first bit string includes the first bit value at the first position; the flip unit 43 is configured to:

[0097]If the first bit value is 1, keep its bit value unchanged with a third probability;

[0098]If the value of the first bit is 0, turn it to 1 with a fourth probability;

[0099]Wherein, at least one of the third probability and the fourth probability is determined according to the second privacy budget.

[0100]According to another implementation manner, for the first bit value of the above-mentioned first position, the inversion unit 43 is configured to:

[0101]If the first position is an even number, set its flip value to 1 with a fifth probability;

[0102]If the first position is an odd number, set its flip value to 1 with a sixth probability;

[0103]Wherein, the fifth probability and the sixth probability are different values ​​respectively determined according to the second privacy budget and the length of the first bit string.

[01...

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 for protecting privacy in a distributed system, the distributed system at least comprises a plurality of data platforms, andthe method can be executed by any data platform and comprises the following steps: firstly, utilizing a local sample set to train a service prediction model to obtain a floating point numerical valueof each weight parameter; converting the floating point numerical value of each weight parameter into a binary bit value by using a first randomization algorithm to obtain a first bit string; then, randomly overturning the bit value of each position in the first bit string by using a second randomization algorithm to obtain a second bit string. Thus, a second bit string may be provided as a localized training result for each weight parameter of the traffic prediction model.

Description

Technical field[0001]One or more embodiments of this specification relate to the field of machine learning, and in particular to a model training method and device for protecting privacy in a distributed system.Background technique[0002]The rapid development of machine learning has enabled various machine learning models to be applied in a variety of business scenarios. Since the prediction performance of a model depends on the richness and availability of training samples, in order to obtain a business prediction model with better performance, it is often necessary to comprehensively use training data from multiple platforms to jointly train the model.[0003]Specifically, in a scenario where data is vertically distributed, multiple platforms may each have different characteristic data of the same batch of business objects. For example, in a business classification analysis scenario based on machine learning, the electronic payment platform owns the merchant's transaction flow data, ...

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): G06F21/60G06N3/04G06N3/08
CPCG06F21/604G06N3/08G06N3/045G06N3/044
Inventor 熊涛
Owner ALIPAY (HANGZHOU) INFORMATION 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