Title:
METHOD AND APPARATUS FOR TRAINING A LEARNING MACHINE
Document Type and Number:
WIPO Patent Application WO/2017/156791
Kind Code:
A1
Abstract:
The disclosure relates to a method and apparatus for training a learning machine, wherein the apparatus includes : a broadcasting module for broadcasting an initial global model for a training cycle to a plurality of worker nodes; a receiving module for receiving a plurality of updated local models from the plurality of worker nodes, wherein each updated local model is generated by one of the plurality of worker nodes independently based on a data split assigned to the worker node and the initial global model for the training cycle; an aggregating module for aggregating the plurality of updated local models to obtain an aggregated model; and a generating module for generating an updated global model for the training cycle based at least on the aggregated model and historical information which is obtained from a preceding training cycle.
Inventors:
CHEN KAI (US)
HUO QIANG (US)
HUO QIANG (US)
Application Number:
PCT/CN2016/076808
Publication Date:
September 21, 2017
Filing Date:
March 18, 2016
Export Citation:
Assignee:
MICROSOFT TECHNOLOGY LICENSING LLC (US)
CHEN KAI (CN)
HUO QIANG (CN)
CHEN KAI (CN)
HUO QIANG (CN)
International Classes:
G06N20/00
Foreign References:
US8027938B1 | 2011-09-27 | |||
CN103745225A | 2014-04-23 | |||
US20110295567A1 | 2011-12-01 | |||
US20160073969A1 | 2016-03-17 | |||
US9235813B1 | 2016-01-12 | |||
US20150134576A1 | 2015-05-14 |
Other References:
HANG SU ET AL., EXPERIMENTS ON PARALLEL TRAINING OF DEEP NEURAL NETWORK USING MODEL AVERAGING
See also references of EP 3430526A4
See also references of EP 3430526A4
Attorney, Agent or Firm:
NTD PATENT & TRADEMARK AGENCY LIMITED (CN)
Download PDF: