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

Distributed system and method for performing machine learning

A distributed system and machine learning technology, applied in machine learning, instruments, computer parts, etc., can solve the problem of large storage overhead of parameter servers, and achieve the effect of reducing the amount of calculation, network transmission overhead, and storage overhead.

Active Publication Date: 2020-04-07
THE FOURTH PARADIGM BEIJING TECH CO LTD
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Exemplary embodiments of the present invention are to provide a distributed system and method for performing machine learning, so as to solve the problem of excessive storage overhead of the parameter server when performing operations of multiple machine learning models at the same time

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
  • Distributed system and method for performing machine learning
  • Distributed system and method for performing machine learning
  • Distributed system and method for performing machine learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0031] Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like numerals refer to like parts throughout. The embodiments are described below in order to explain the present invention by referring to the figures.

[0032] The distributed system for performing machine learning involved in the exemplary embodiments of the present invention may be composed of a parameter server and a plurality of computing devices, wherein the parameter server maintains parameters of a plurality of machine learning models by interacting with each computing device, Multiple computing devices perform training and / or estimation on the multiple machine learning models in parallel. It should be noted that the computing device and / or parameter server mentioned here are all defined by the processing executed or the functions realized, which may refer to a physical entity or a virtual entity, for example, a 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

A distributed system and method for performing machine learning are provided. The distributed system includes: a parameter server for maintaining parameters of multiple machine learning models, wherein the parameters are in the form of key-value pairs, and the parameter server saves the parameters with a plurality of key-value pairs for the same key; and a plurality of computing devices configured to execute algorithms on the plurality of machine learning models in parallel. According to the distributed system and method thereof, it is possible to effectively reduce the storage overhead of the parameter server when training and / or estimating multiple machine learning models at the same time.

Description

technical field [0001] Exemplary embodiments of the present invention generally relate to the field of artificial intelligence, and more particularly, to a distributed system for performing machine learning and a method for performing machine learning using the distributed system. Background technique [0002] With the rapid growth of data scale, machine learning is widely used in various fields to mine the value of data. However, in order to perform machine learning, the memory of general physical machines is far from enough. For this reason, in practice, it is often necessary to use a distributed machine learning system to complete the training of machine learning models or corresponding predictions. In the existing distributed machine learning system, the training or estimation of the same machine learning model is usually performed in parallel by multiple computing devices, and the parameters of the machine learning model are stored by the parameter server, and each comp...

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 Patents(China)
IPC IPC(8): G06F16/22G06K9/62G06N20/00
CPCG06F16/22G06N20/00G06F18/214G06F16/2237
Inventor 戴文渊杨强陈雨强刘一鸣石光川
Owner THE FOURTH PARADIGM BEIJING 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