MACHINE LEARNING BASED NETWORK MODEL CONSTRUCTION METHOD AND APPARATUS

Title:

MACHINE LEARNING BASED NETWORK MODEL CONSTRUCTION METHOD AND APPARATUS

Document Type and Number:

WIPO Patent Application WO/2017/206936

Kind Code:

A1

Abstract:

A machine learning based network model construction method and apparatus. The method comprises: acquiring a data processing flow of an original network model and a reference data set generated by the original network model in the data processing flow (S101); according to the data processing flow and the reference data set, hierarchically constructing at least one sub-network (S102); using the reference data set to perform optimization training on the at least one sub-network (S103); and combining and processing the at least one sub-network after the optimization training to form a target network model (S104). The technical solution can simplify a construction process of a network model, thereby improving the model construction efficiency.

Inventors:

ZHENG BO (CN)
LIU ZHIBIN (CN)
LIU RIJIA (CN)
CHEN QIAN (CN)

Application Number:

PCT/CN2017/086917

Publication Date:

December 07, 2017

Filing Date:

June 02, 2017

Export Citation:

Click for automatic bibliography generation Help

Assignee:

TENCENT TECH SHENZHEN CO LTD (CN)

International Classes:

G06N3/08

Foreign References:

CN106096727A	2016-11-09
CN104751227A	2015-07-01
CN103729459A	2014-04-16
CN101814160A	2010-08-25
EP2533176A1	2012-12-12

Other References: