Discriminative training in machine learning


Ontology type: sgo:Patent     


Patent Info

DATE

N/A

AUTHORS

Peng Xu , Ioannis Tsochandaridis

ABSTRACT

Systems, methods, and apparatuses including computer program products for machine learning are provided. A method is provided that includes distributing a parameterized model to each worker of a hierarchy of workers, the parameterized model including a plurality of feature functions and corresponding model parameters, processing a portion of training data at each worker of the plurality of workers according to the parameterized model to calculate updates to model parameters, for each worker at a lowest level of the hierarchy of workers, sending the calculated updates to a next higher level worker, for each other worker in the hierarchy of workers, combining updates of the respective worker with updates received from one or more lower level workers, collecting all updates from the workers at a master to generate real updates to the model parameters, and generating an updated model using the real updates to the model parameters. More... »

Related SciGraph Publications

JSON-LD is the canonical representation for SciGraph data.

TIP: You can open this SciGraph record using an external JSON-LD service: JSON-LD Playground Google SDTT

[
  {
    "@context": "https://springernature.github.io/scigraph/jsonld/sgcontext.json", 
    "about": [
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/2746", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "type": "DefinedTerm"
      }
    ], 
    "author": [
      {
        "name": "Peng Xu", 
        "type": "Person"
      }, 
      {
        "name": "Ioannis Tsochandaridis", 
        "type": "Person"
      }
    ], 
    "citation": [
      {
        "id": "sg:pub.10.1007/bf01589116", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1022481421", 
          "https://doi.org/10.1007/bf01589116"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/12.543711", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1061088462"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/cvpr.1988.196212", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1086178452"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/mwscas.1994.519395", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1093323172"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/lics.2004.1319633", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1095624080"
        ], 
        "type": "CreativeWork"
      }
    ], 
    "description": "

Systems, methods, and apparatuses including computer program products for machine learning are provided. A method is provided that includes distributing a parameterized model to each worker of a hierarchy of workers, the parameterized model including a plurality of feature functions and corresponding model parameters, processing a portion of training data at each worker of the plurality of workers according to the parameterized model to calculate updates to model parameters, for each worker at a lowest level of the hierarchy of workers, sending the calculated updates to a next higher level worker, for each other worker in the hierarchy of workers, combining updates of the respective worker with updates received from one or more lower level workers, collecting all updates from the workers at a master to generate real updates to the model parameters, and generating an updated model using the real updates to the model parameters.

", "id": "sg:patent.US-8027938-B1", "keywords": [ "machine", "method", "apparatus", "computer", "parameterized model", "worker", "hierarchy", "plurality", "feature", "corresponding model", "processing", "portion", "training data", "update", "model parameter", "low level", "high level", "updated model" ], "name": "Discriminative training in machine learning", "recipient": [ { "id": "https://www.grid.ac/institutes/grid.420451.6", "type": "Organization" } ], "sameAs": [ "https://app.dimensions.ai/details/patent/US-8027938-B1" ], "sdDataset": "patents", "sdDatePublished": "2019-03-07T15:36", "sdLicense": "https://scigraph.springernature.com/explorer/license/", "sdPublisher": { "name": "Springer Nature - SN SciGraph project", "type": "Organization" }, "sdSource": "s3://com.uberresearch.data.dev.patents-pipeline/full_run_10/sn-export/5eb3e5a348d7f117b22cc85fb0b02730/0000100128-0000348334/json_export_c290c18d.jsonl", "type": "Patent" } ]
 

Download the RDF metadata as:  json-ld nt turtle xml License info

HOW TO GET THIS DATA PROGRAMMATICALLY:

JSON-LD is a popular format for linked data which is fully compatible with JSON.

curl -H 'Accept: application/ld+json' 'https://scigraph.springernature.com/patent.US-8027938-B1'

N-Triples is a line-based linked data format ideal for batch operations.

curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/patent.US-8027938-B1'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/patent.US-8027938-B1'

RDF/XML is a standard XML format for linked data.

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/patent.US-8027938-B1'


 

This table displays all metadata directly associated to this object as RDF triples.

59 TRIPLES      14 PREDICATES      36 URIs      25 LITERALS      2 BLANK NODES

Subject Predicate Object
1 sg:patent.US-8027938-B1 schema:about anzsrc-for:2746
2 schema:author N960135cf428044b0b797d34fd93a951e
3 schema:citation sg:pub.10.1007/bf01589116
4 https://doi.org/10.1109/12.543711
5 https://doi.org/10.1109/cvpr.1988.196212
6 https://doi.org/10.1109/lics.2004.1319633
7 https://doi.org/10.1109/mwscas.1994.519395
8 schema:description <p num="p-0001">Systems, methods, and apparatuses including computer program products for machine learning are provided. A method is provided that includes distributing a parameterized model to each worker of a hierarchy of workers, the parameterized model including a plurality of feature functions and corresponding model parameters, processing a portion of training data at each worker of the plurality of workers according to the parameterized model to calculate updates to model parameters, for each worker at a lowest level of the hierarchy of workers, sending the calculated updates to a next higher level worker, for each other worker in the hierarchy of workers, combining updates of the respective worker with updates received from one or more lower level workers, collecting all updates from the workers at a master to generate real updates to the model parameters, and generating an updated model using the real updates to the model parameters.</p>
9 schema:keywords apparatus
10 computer
11 corresponding model
12 feature
13 hierarchy
14 high level
15 low level
16 machine
17 method
18 model parameter
19 parameterized model
20 plurality
21 portion
22 processing
23 training data
24 update
25 updated model
26 worker
27 schema:name Discriminative training in machine learning
28 schema:recipient https://www.grid.ac/institutes/grid.420451.6
29 schema:sameAs https://app.dimensions.ai/details/patent/US-8027938-B1
30 schema:sdDatePublished 2019-03-07T15:36
31 schema:sdLicense https://scigraph.springernature.com/explorer/license/
32 schema:sdPublisher N4436bf4d33bf4c43aa7307b7ab7c3a2c
33 sgo:license sg:explorer/license/
34 sgo:sdDataset patents
35 rdf:type sgo:Patent
36 N4436bf4d33bf4c43aa7307b7ab7c3a2c schema:name Springer Nature - SN SciGraph project
37 rdf:type schema:Organization
38 N569c3fc4f3174ea38f8292a70796a539 rdf:first N8153ee5c74ac4fb1884b4571fa68fe1a
39 rdf:rest rdf:nil
40 N8153ee5c74ac4fb1884b4571fa68fe1a schema:name Ioannis Tsochandaridis
41 rdf:type schema:Person
42 N960135cf428044b0b797d34fd93a951e rdf:first Nc9e5861bafc044cf9c41b658bbe68ca0
43 rdf:rest N569c3fc4f3174ea38f8292a70796a539
44 Nc9e5861bafc044cf9c41b658bbe68ca0 schema:name Peng Xu
45 rdf:type schema:Person
46 anzsrc-for:2746 schema:inDefinedTermSet anzsrc-for:
47 rdf:type schema:DefinedTerm
48 sg:pub.10.1007/bf01589116 schema:sameAs https://app.dimensions.ai/details/publication/pub.1022481421
49 https://doi.org/10.1007/bf01589116
50 rdf:type schema:CreativeWork
51 https://doi.org/10.1109/12.543711 schema:sameAs https://app.dimensions.ai/details/publication/pub.1061088462
52 rdf:type schema:CreativeWork
53 https://doi.org/10.1109/cvpr.1988.196212 schema:sameAs https://app.dimensions.ai/details/publication/pub.1086178452
54 rdf:type schema:CreativeWork
55 https://doi.org/10.1109/lics.2004.1319633 schema:sameAs https://app.dimensions.ai/details/publication/pub.1095624080
56 rdf:type schema:CreativeWork
57 https://doi.org/10.1109/mwscas.1994.519395 schema:sameAs https://app.dimensions.ai/details/publication/pub.1093323172
58 rdf:type schema:CreativeWork
59 https://www.grid.ac/institutes/grid.420451.6 schema:Organization
 




Preview window. Press ESC to close (or click here)


...