Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
METHOD AND APPARATUS FOR DISTRIBUTED MACHINE LEARNING SYSTEM
Document Type and Number:
WIPO Patent Application WO/2018/223330
Kind Code:
A1
Abstract:
A server for facilitating dependency-based execution of a set of machine learning tasks is provided. The server stores data representing storage elements in a machine learning model and connections between the storage elements. The server also comprises a task module configured to receive processing assignments for some of the storage elements, and determine a set of machine learning tasks and dependencies among the set of machine learning. The server also comprises a scheduler module configured to determine a resolution status for each of the dependencies, and determine, based on the resolution status, a sequence of execution of the set of machine learning tasks. The server also comprises one or more hardware processor configured to execute a set of machine learning tasks based on the determined sequence.

Inventors:
ZOU YOUZHI (US)
ZHANG JIAXING (CN)
CUI XIAOYUAN (US)
LI XIAOLONG (US)
YUAN QI (CN)
Application Number:
CN2017/087570
Publication Date:
December 13, 2018
Filing Date:
June 08, 2017
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
ALIBABA GROUP HOLDING LTD (CN)
International Classes:
G06F19/24
Foreign References:
US20110131163A12011-06-02
US20130290223A12013-10-31
US20140222730A12014-08-07
Other References:
HECKERMAN, DAVID ET AL.: "Dependency Networks for Inference, Collaborative Filtering,and Data Visualization", JOURNAL OF MACHINE LEARNING RESEARCH, 31 December 2000 (2000-12-31), pages 49 - 75, XP058153209, ISSN: 1532-4435
Attorney, Agent or Firm:
BEIJING TSINGYUANHUI INTELLECTUAL PROPERTY LAW FIRM (Haidian District, Beijing 0, CN)
Download PDF: