Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Weight training method and device, computer equipment and storage medium

A training method and weight technology, applied in the field of machine learning, can solve the problem of low efficiency of training weight

Active Publication Date: 2020-08-25
GUANGZHOU BAIGUOYUAN INFORMATION TECH CO LTD
View PDF5 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] Embodiments of the present invention provide a weight training method, device, computer equipment, and storage medium to solve the problem of low efficiency of training weights when optimizing model weights with multiple objectives

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
  • Weight training method and device, computer equipment and storage medium
  • Weight training method and device, computer equipment and storage medium
  • Weight training method and device, computer equipment and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0102] figure 1 It is a flow chart of a weight training method provided by Embodiment 1 of the present invention. This embodiment can be applied offline to use evaluation indicators to reflect the performance of multi-objective weights after going online, so as to train weights. This method can be implemented by a weight training device To perform, the weight training device can be implemented by software and / or hardware, and can be configured in computer equipment, such as servers, workstations, personal computers, etc., the method specifically includes the following steps:

[0103] S101. Obtain a business model.

[0104] In this embodiment, the business model and its parameters can be pre-trained, and the business model and its parameters are stored in the database. When training weights, the business model can be read from the database and its parameters loaded.

[0105] Wherein, the business model is used to calculate the probability that the user performs multiple (two o...

Embodiment 2

[0193] image 3 It is a schematic structural diagram of a weight training device provided in Embodiment 2 of the present invention. The device may specifically include the following modules:

[0194] A business model acquiring module 301, configured to acquire a business model, and the business model is used to calculate the probability that a user performs multiple target actions on business data;

[0195] A weight configuration module 302, configured to configure a weight for the target behavior;

[0196] The offline evaluation index calculation module 303 is used to calculate the evaluation index presented by the target behavior to push the business data to the user under the weight, as an offline evaluation index;

[0197] An index state identification module 304, configured to identify the state of the offline evaluation index;

[0198] A weight adjustment module 305, configured to adjust the weight according to the evaluation index if the state is that the offline eval...

Embodiment 3

[0236] Figure 4 It is a schematic structural diagram of a computer device provided by Embodiment 3 of the present invention. Such as Figure 4 As shown, the computer device includes a processor 400, a memory 401, a communication module 402, an input device 403 and an output device 404; the number of processors 400 in the computer device may be one or more, Figure 4 Take a processor 400 as an example; the processor 400, memory 401, communication module 402, input device 403 and output device 404 in the computer equipment can be connected by bus or other methods, Figure 4 Take connection via bus as an example.

[0237] The memory 401, as a computer-readable storage medium, can be used to store software programs, computer-executable programs and modules, such as modules corresponding to the weight training method in this embodiment (for example, such as image 3 The business model acquisition module 301, the weight configuration module 302, the offline evaluation indicator ...

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 weight training method and device, computer equipment and a storage medium. The method comprises the steps of acquiring a service model; configuring a weight for the target behavior; calculating an evaluation index presented by pushing the business data to the user by the target behavior under the weight, and taking the evaluation index as an offline evaluation index; identifying the state of the offline evaluation index; if the state is that the offline evaluation index is not converged, adjusting the weight according to the evaluation index, and returning to calculate the evaluation index presented by the target behavior pushing the service data to the user under the weight as the offline evaluation index; and if the state is that the offline evaluation index is converged, determining that the weight training of the service model is completed. According to the method of the invention, in an offline state, parameter searching (namely weightsetting) is guided through the evaluation indexes, the parameter searching direction is defined, the precision of the weight in offline training is improved, the frequency of weight adjustment according to online conditions can be reduced, a large amount of time and manpower are saved, and thus the efficiency is improved.

Description

technical field [0001] Embodiments of the present invention relate to machine learning techniques, and in particular, to a weight training method, device, computer equipment, and storage medium. Background technique [0002] In business scenarios such as information retrieval and information recommendation, the pre-trained model usually recalls business data, selects appropriate business data from all business data and sends it to the user, and the user operates on the business data, such as clicking, liking, commenting, sharing , attention, etc. [0003] In this process, the CTR (Click Through Rate) is usually used as the target for optimization. For example, for multimedia business scenarios, the optimization goal is the short video that the user clicks to send. This method leads to the return to The user's information pays more attention to the information directly displayed to the user, such as the title and cover of the business data, so that it is easier for the user ...

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): G06N3/04G06N3/08G06N20/00
CPCG06N3/084G06N20/00G06N3/045
Inventor 徐宣宏
Owner GUANGZHOU BAIGUOYUAN 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
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