Title:
METHOD AND DEVICE FOR PUSHING OBJECT TO USER BASED ON REINFORCEMENT LEARNING MODEL
Document Type and Number:
WIPO Patent Application WO/2020/220757
Kind Code:
A1
Abstract:
A method and device for determining a push object list for a user based on a reinforcement learning model. The method comprises: for each group of object lists, obtaining the ith state feature vector (S202); inputting the ith state feature vector into the reinforcement learning model to enable the reinforcement learning model to output a weight vector corresponding to the ith state feature vector (S204); obtaining sorting feature vectors of objects in a candidate object set corresponding to the group of object lists (S206); calculating scores of the objects in the candidate object set on the basis of a point product of the sorting feature vectors of the objects in the candidate object set and the weight vector (S208); and for the M groups of object lists, determining updated M groups of object lists on the basis of the scores of the objects in M candidate object sets corresponding to the M groups of object lists (S210), wherein each group of object lists in the updated M groups of object lists comprises i objects.
Inventors:
CHEN CEN (CN)
HU XU (CN)
FU CHILIN (CN)
ZHANG XIAOLU (CN)
HU XU (CN)
FU CHILIN (CN)
ZHANG XIAOLU (CN)
Application Number:
PCT/CN2020/071699
Publication Date:
November 05, 2020
Filing Date:
January 13, 2020
Export Citation:
Assignee:
ALIBABA GROUP HOLDING LTD (CN)
International Classes:
G06F16/9535
Foreign References:
| CN110263245A | 2019-09-20 | |||
| CN108304440A | 2018-07-20 | |||
| CN104869464A | 2015-08-26 | |||
| CN108805594A | 2018-11-13 | |||
| US20140297476A1 | 2014-10-02 |
Attorney, Agent or Firm:
BEIJING BESTIPR INTELLECTUAL PROPERTY LAW CORPORATION (CN)
Download PDF:
Previous Patent: NETWORK CONFIGURATION METHOD AND APPARATUS
Next Patent: METHOD FOR DETECTING ABNORMAL TRANSACTION NODE, AND DEVICE
Next Patent: METHOD FOR DETECTING ABNORMAL TRANSACTION NODE, AND DEVICE