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

Method for training recommendation probability prediction model and recommendation probability prediction method and device

A probabilistic prediction and model technology, applied in the field of artificial intelligence, can solve problems such as the difference in the output results of multiple neural networks, the impact of the prediction accuracy of the information to be predicted, and the convergence of the output results of multiple neural networks, so as to improve the prediction performance and improve the prediction performance. The effect of accuracy

Active Publication Date: 2021-08-13
TENCENT TECH (SHENZHEN) CO LTD
View PDF6 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the current multi-task learning model usually has the problem of the convergence of the output results of multiple neural networks, that is, the output results of multiple neural networks are very small, so that the prediction performance of the multi-task learning model after training is not high, which will treat The forecast accuracy of the forecast information is affected

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
  • Method for training recommendation probability prediction model and recommendation probability prediction method and device
  • Method for training recommendation probability prediction model and recommendation probability prediction method and device
  • Method for training recommendation probability prediction model and recommendation probability prediction method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0080] In order to make the purpose, technical solution and advantages of the application clearer, the application will be further described in detail below in conjunction with the accompanying drawings. Apparently, the described embodiments are only some of the embodiments of the application, not all of them. Based on the embodiments in this application, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the scope of protection of this application.

[0081] In order to facilitate those skilled in the art to better understand the technical solutions of the present application, terms involved in the present application are introduced below.

[0082] Multi-task learning model: A model learns multiple tasks at the same time. For example, in the video recommendation scenario: a recommendation model not only learns the click probability prediction task, but also learns the playback duration prediction task, and learns the...

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 invention provides a method for training a recommendation probability prediction model and a recommendation probability prediction method and device, and relates to the technical field of artificial intelligence, and the method comprises the steps: inputting sample data obtained from a sample data set into the recommendation probability prediction model in the multi-round iterative training process of the recommendation probability prediction model, obtaining a prediction recommendation result corresponding to the sample data; obtaining a corresponding basic loss value according to a predicted recommendation result and an actual recommendation result corresponding to the sample data; obtaining a corresponding target loss value based on the basic loss value and the first adjustment value, wherein the first adjustment value is used for representing the total difference degree of output results, obtained based on the corresponding sample data, of every two task networks, and the first adjustment value is in negative correlation with the target loss value; and according to the target loss value, performing parameter adjustment on the recommendation probability prediction model to improve the prediction performance of the trained recommendation probability prediction model so as to improve the prediction accuracy of the to-be-predicted information.

Description

technical field [0001] The present application relates to the technical field of artificial intelligence, and in particular to a method for training a recommendation probability prediction model, a recommendation probability prediction method and a device. Background technique [0002] With the continuous development of Internet technology, various network information emerges in an endless stream, such as articles, videos, pictures, products, advertisements, etc., making information recommendation systems widely used. In practical applications, when recommending articles, videos, pictures and other information to the target object through the information recommendation system, the purpose is to achieve the following recommendation effect: the target object clicks on the recommended information, then reads the entire content of the recommended information, and likes it , forwarding, commenting and other interactions. [0003] In related technologies, in order to achieve the ...

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
IPC IPC(8): G06F16/9535G06F16/735G06N3/08G06N3/04
CPCG06F16/9535G06F16/735G06N3/08G06N3/045
Inventor 伍海洋
Owner TENCENT TECH (SHENZHEN) 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