Patent · US Active

Training a model using parameter server shards

US8768870B1 · kind B1 · utility

52Cited by
0References
23Claims
0Family size

Assignee

Inventors

Key dates

Filing dateAug 15, 2013
Grant dateJul 1, 2014
Priority date
Expiry dateAug 15, 2033

Classification

  • Technology area (CPC G)Physics
  • CPC primaryG06N7/01
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a model using parameter server shards. One of the methods includes receiving, at a parameter server shard configured to maintain values of a disjoint partition of the parameters of the model, a succession of respective requests for parameter values from each of a plurality of replicas of the model; in response to each request, downloading a current value of each requested parameter to the replica from which the request was received; receiving a succession of uploads, each upload including respective delta values for each of the parameters in the partition maintained by the shard; and updating values of the parameters in the partition maintained by the parameter server shard repeatedly based on the uploads of delta values to generate current parameter values.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.