Multimedia recommendation model training method and device, server and storage medium

The invention relates to a multimedia recommendation model training method and device, a server and a storage medium, and belongs to the technical field of deep learning, and the method comprises the steps: carrying out the iterative training of a plurality of multimedia recommendation models, deter...

Full description

Saved in:
Bibliographic Details
Main Authors JIA JIYUAN, LI JIXIANG, YANG SEN
Format Patent
LanguageChinese
English
Published 08.06.2021
Subjects
CALCULATING
COMPUTING
COUNTING
ELECTRIC DIGITAL DATA PROCESSING
HANDLING RECORD CARRIERS
PHYSICS
PRESENTATION OF DATA
RECOGNITION OF DATA
RECORD CARRIERS
Online AccessGet full text

Cover

More Information
Summary:The invention relates to a multimedia recommendation model training method and device, a server and a storage medium, and belongs to the technical field of deep learning, and the method comprises the steps: carrying out the iterative training of a plurality of multimedia recommendation models, determining a correlation model corresponding to each multimedia recommendation model, and respectively determining target model parameters of each multimedia recommendation model based on the model parameters of each multimedia recommendation model and the model parameters of the corresponding association model. According to the technical scheme provided by the embodiment of the invention, the association model determined by each multimedia recommendation model each time is determined based on the association model determined last time, so that the model parameters among the multimedia recommendation models can be ensured to be fused and interacted as far as possible; according to the method and device, parameter optim
Bibliography:Application Number: CN202110120344