Dual subgradient algorithms for large-scale nonsmooth learning problems View Full Text


Ontology type: schema:ScholarlyArticle     


Article Info

DATE

2014-12

AUTHORS

Bruce Cox, Anatoli Juditsky, Arkadi Nemirovski

ABSTRACT

“Classical” First Order (FO) algorithms of convex optimization, such as Mirror Descent algorithm or Nesterov’s optimal algorithm of smooth convex optimization, are well known to have optimal (theoretical) complexity estimates which do not depend on the problem dimension. However, to attain the optimality, the domain of the problem should admit a “good proximal setup”. The latter essentially means that (1) the problem domain should satisfy certain geometric conditions of “favorable geometry”, and (2) the practical use of these methods is conditioned by our ability to compute at a moderate cost proximal transformation at each iteration. More often than not these two conditions are satisfied in optimization problems arising in computational learning, what explains why proximal type FO methods recently became methods of choice when solving various learning problems. Yet, they meet their limits in several important problems such as multi-task learning with large number of tasks, where the problem domain does not exhibit favorable geometry, and learning and matrix completion problems with nuclear norm constraint, when the numerical cost of computing proximal transformation becomes prohibitive in large-scale problems. We propose a novel approach to solving nonsmooth optimization problems arising in learning applications where Fenchel-type representation of the objective function is available. The approach is based on applying FO algorithms to the dual problem and using the accuracy certificates supplied by the method to recover the primal solution. While suboptimal in terms of accuracy guaranties, the proposed approach does not rely upon “good proximal setup” for the primal problem but requires the problem domain to admit a Linear Optimization oracle—the ability to efficiently maximize a linear form on the domain of the primal problem. More... »

PAGES

143-180

References to SciGraph publications

Identifiers

URI

http://scigraph.springernature.com/pub.10.1007/s10107-013-0725-1

DOI

http://dx.doi.org/10.1007/s10107-013-0725-1

DIMENSIONS

https://app.dimensions.ai/details/publication/pub.1029579235


Indexing Status Check whether this publication has been indexed by Scopus and Web Of Science using the SN Indexing Status Tool
Incoming Citations Browse incoming citations for this publication using opencitations.net

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/2008/0103", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Numerical and Computational Mathematics", 
        "type": "DefinedTerm"
      }, 
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/01", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Mathematical Sciences", 
        "type": "DefinedTerm"
      }
    ], 
    "author": [
      {
        "affiliation": {
          "alternateName": "United States Department of the Air Force", 
          "id": "https://www.grid.ac/institutes/grid.495456.f", 
          "name": [
            "US Air Force, Washington, DC, USA"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Cox", 
        "givenName": "Bruce", 
        "id": "sg:person.010340346467.73", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.010340346467.73"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Laboratoire Jean Kuntzmann", 
          "id": "https://www.grid.ac/institutes/grid.464181.e", 
          "name": [
            "LJK, Universit\u00e9 J. Fourier, B.P. 53, 38041, Grenoble Cedex 9, France"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Juditsky", 
        "givenName": "Anatoli", 
        "id": "sg:person.016013144233.55", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016013144233.55"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Georgia Institute of Technology", 
          "id": "https://www.grid.ac/institutes/grid.213917.f", 
          "name": [
            "Georgia Institute of Technology, 30332, Atlanta, GA, USA"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Nemirovski", 
        "givenName": "Arkadi", 
        "id": "sg:person.011304472045.51", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.011304472045.51"
        ], 
        "type": "Person"
      }
    ], 
    "citation": [
      {
        "id": "https://doi.org/10.1214/11-aos944", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1002363869"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1016/0022-247x(78)90137-3", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1009751684"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1016/0041-5553(67)90040-7", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1010322999"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/s10107-007-0149-x", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1016713895", 
          "https://doi.org/10.1007/s10107-007-0149-x"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/s10107-007-0149-x", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1016713895", 
          "https://doi.org/10.1007/s10107-007-0149-x"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/s10107-004-0553-4", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1021331750", 
          "https://doi.org/10.1007/s10107-004-0553-4"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/s10107-004-0552-5", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1030642071", 
          "https://doi.org/10.1007/s10107-004-0552-5"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1145/1273496.1273499", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1031215462"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/s11228-010-0147-7", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1032025040", 
          "https://doi.org/10.1007/s11228-010-0147-7"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/s11228-010-0147-7", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1032025040", 
          "https://doi.org/10.1007/s11228-010-0147-7"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/bf01585555", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1038375449", 
          "https://doi.org/10.1007/bf01585555"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/s10107-006-0034-z", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1042069345", 
          "https://doi.org/10.1007/s10107-006-0034-z"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/s10107-006-0034-z", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1042069345", 
          "https://doi.org/10.1007/s10107-006-0034-z"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1002/nav.3800030109", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1047094257"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/978-1-4419-8853-9", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1049823794", 
          "https://doi.org/10.1007/978-1-4419-8853-9"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/978-1-4419-8853-9", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1049823794", 
          "https://doi.org/10.1007/978-1-4419-8853-9"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1017/s096249291300007x", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1054905297"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1137/0803026", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1062854211"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1137/s1052623403425629", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1062883356"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1287/moor.1090.0427", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1064723060"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.24033/bsmf.1625", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1083660582"
        ], 
        "type": "CreativeWork"
      }
    ], 
    "datePublished": "2014-12", 
    "datePublishedReg": "2014-12-01", 
    "description": "\u201cClassical\u201d First Order (FO) algorithms of convex optimization, such as Mirror Descent algorithm or Nesterov\u2019s optimal algorithm of smooth convex optimization, are well known to have optimal (theoretical) complexity estimates which do not depend on the problem dimension. However, to attain the optimality, the domain of the problem should admit a \u201cgood proximal setup\u201d. The latter essentially means that (1) the problem domain should satisfy certain geometric conditions of \u201cfavorable geometry\u201d, and (2) the practical use of these methods is conditioned by our ability to compute at a moderate cost proximal transformation at each iteration. More often than not these two conditions are satisfied in optimization problems arising in computational learning, what explains why proximal type FO methods recently became methods of choice when solving various learning problems. Yet, they meet their limits in several important problems such as multi-task learning with large number of tasks, where the problem domain does not exhibit favorable geometry, and learning and matrix completion problems with nuclear norm constraint, when the numerical cost of computing proximal transformation becomes prohibitive in large-scale problems. We propose a novel approach to solving nonsmooth optimization problems arising in learning applications where Fenchel-type representation of the objective function is available. The approach is based on applying FO algorithms to the dual problem and using the accuracy certificates supplied by the method to recover the primal solution. While suboptimal in terms of accuracy guaranties, the proposed approach does not rely upon \u201cgood proximal setup\u201d for the primal problem but requires the problem domain to admit a Linear Optimization oracle\u2014the ability to efficiently maximize a linear form on the domain of the primal problem.", 
    "genre": "research_article", 
    "id": "sg:pub.10.1007/s10107-013-0725-1", 
    "inLanguage": [
      "en"
    ], 
    "isAccessibleForFree": false, 
    "isPartOf": [
      {
        "id": "sg:journal.1047630", 
        "issn": [
          "0025-5610", 
          "1436-4646"
        ], 
        "name": "Mathematical Programming", 
        "type": "Periodical"
      }, 
      {
        "issueNumber": "1-2", 
        "type": "PublicationIssue"
      }, 
      {
        "type": "PublicationVolume", 
        "volumeNumber": "148"
      }
    ], 
    "name": "Dual subgradient algorithms for large-scale nonsmooth learning problems", 
    "pagination": "143-180", 
    "productId": [
      {
        "name": "readcube_id", 
        "type": "PropertyValue", 
        "value": [
          "8ce90f27d91fe062c9b0d9c91e7a1f51d77e3afcbc5b76e1bfffe1ad1c7f3151"
        ]
      }, 
      {
        "name": "doi", 
        "type": "PropertyValue", 
        "value": [
          "10.1007/s10107-013-0725-1"
        ]
      }, 
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "pub.1029579235"
        ]
      }
    ], 
    "sameAs": [
      "https://doi.org/10.1007/s10107-013-0725-1", 
      "https://app.dimensions.ai/details/publication/pub.1029579235"
    ], 
    "sdDataset": "articles", 
    "sdDatePublished": "2019-04-11T01:11", 
    "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
    "sdPublisher": {
      "name": "Springer Nature - SN SciGraph project", 
      "type": "Organization"
    }, 
    "sdSource": "s3://com-uberresearch-data-dimensions-target-20181106-alternative/cleanup/v134/2549eaecd7973599484d7c17b260dba0a4ecb94b/merge/v9/a6c9fde33151104705d4d7ff012ea9563521a3ce/jats-lookup/v90/0000000001_0000000264/records_8697_00000532.jsonl", 
    "type": "ScholarlyArticle", 
    "url": "http://link.springer.com/10.1007%2Fs10107-013-0725-1"
  }
]
 

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/pub.10.1007/s10107-013-0725-1'

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

curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1007/s10107-013-0725-1'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/s10107-013-0725-1'

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

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/s10107-013-0725-1'


 

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

139 TRIPLES      21 PREDICATES      44 URIs      19 LITERALS      7 BLANK NODES

Subject Predicate Object
1 sg:pub.10.1007/s10107-013-0725-1 schema:about anzsrc-for:01
2 anzsrc-for:0103
3 schema:author N50486dbb46064346b2c8bf1e8ac46155
4 schema:citation sg:pub.10.1007/978-1-4419-8853-9
5 sg:pub.10.1007/bf01585555
6 sg:pub.10.1007/s10107-004-0552-5
7 sg:pub.10.1007/s10107-004-0553-4
8 sg:pub.10.1007/s10107-006-0034-z
9 sg:pub.10.1007/s10107-007-0149-x
10 sg:pub.10.1007/s11228-010-0147-7
11 https://doi.org/10.1002/nav.3800030109
12 https://doi.org/10.1016/0022-247x(78)90137-3
13 https://doi.org/10.1016/0041-5553(67)90040-7
14 https://doi.org/10.1017/s096249291300007x
15 https://doi.org/10.1137/0803026
16 https://doi.org/10.1137/s1052623403425629
17 https://doi.org/10.1145/1273496.1273499
18 https://doi.org/10.1214/11-aos944
19 https://doi.org/10.1287/moor.1090.0427
20 https://doi.org/10.24033/bsmf.1625
21 schema:datePublished 2014-12
22 schema:datePublishedReg 2014-12-01
23 schema:description “Classical” First Order (FO) algorithms of convex optimization, such as Mirror Descent algorithm or Nesterov’s optimal algorithm of smooth convex optimization, are well known to have optimal (theoretical) complexity estimates which do not depend on the problem dimension. However, to attain the optimality, the domain of the problem should admit a “good proximal setup”. The latter essentially means that (1) the problem domain should satisfy certain geometric conditions of “favorable geometry”, and (2) the practical use of these methods is conditioned by our ability to compute at a moderate cost proximal transformation at each iteration. More often than not these two conditions are satisfied in optimization problems arising in computational learning, what explains why proximal type FO methods recently became methods of choice when solving various learning problems. Yet, they meet their limits in several important problems such as multi-task learning with large number of tasks, where the problem domain does not exhibit favorable geometry, and learning and matrix completion problems with nuclear norm constraint, when the numerical cost of computing proximal transformation becomes prohibitive in large-scale problems. We propose a novel approach to solving nonsmooth optimization problems arising in learning applications where Fenchel-type representation of the objective function is available. The approach is based on applying FO algorithms to the dual problem and using the accuracy certificates supplied by the method to recover the primal solution. While suboptimal in terms of accuracy guaranties, the proposed approach does not rely upon “good proximal setup” for the primal problem but requires the problem domain to admit a Linear Optimization oracle—the ability to efficiently maximize a linear form on the domain of the primal problem.
24 schema:genre research_article
25 schema:inLanguage en
26 schema:isAccessibleForFree false
27 schema:isPartOf N8799b4d9db0d4da580eecb1f9f8c9caf
28 Nc9746db19cd84aa196769b725109ae96
29 sg:journal.1047630
30 schema:name Dual subgradient algorithms for large-scale nonsmooth learning problems
31 schema:pagination 143-180
32 schema:productId N59cd1b5c9e67490892b5dbc0f90865cf
33 N76c0537bbef84489a781a06f0b3521fc
34 N7d26594fe9854a24a814ab31595c51ad
35 schema:sameAs https://app.dimensions.ai/details/publication/pub.1029579235
36 https://doi.org/10.1007/s10107-013-0725-1
37 schema:sdDatePublished 2019-04-11T01:11
38 schema:sdLicense https://scigraph.springernature.com/explorer/license/
39 schema:sdPublisher Nd368eb5a7f3f4ef899527f8ca67da44a
40 schema:url http://link.springer.com/10.1007%2Fs10107-013-0725-1
41 sgo:license sg:explorer/license/
42 sgo:sdDataset articles
43 rdf:type schema:ScholarlyArticle
44 N50486dbb46064346b2c8bf1e8ac46155 rdf:first sg:person.010340346467.73
45 rdf:rest Na039058598554dfdb07218ac67d1c324
46 N59cd1b5c9e67490892b5dbc0f90865cf schema:name dimensions_id
47 schema:value pub.1029579235
48 rdf:type schema:PropertyValue
49 N76c0537bbef84489a781a06f0b3521fc schema:name doi
50 schema:value 10.1007/s10107-013-0725-1
51 rdf:type schema:PropertyValue
52 N7d26594fe9854a24a814ab31595c51ad schema:name readcube_id
53 schema:value 8ce90f27d91fe062c9b0d9c91e7a1f51d77e3afcbc5b76e1bfffe1ad1c7f3151
54 rdf:type schema:PropertyValue
55 N8799b4d9db0d4da580eecb1f9f8c9caf schema:volumeNumber 148
56 rdf:type schema:PublicationVolume
57 Na039058598554dfdb07218ac67d1c324 rdf:first sg:person.016013144233.55
58 rdf:rest Ne5d050690b2e4680bb5437035e30ec08
59 Nc9746db19cd84aa196769b725109ae96 schema:issueNumber 1-2
60 rdf:type schema:PublicationIssue
61 Nd368eb5a7f3f4ef899527f8ca67da44a schema:name Springer Nature - SN SciGraph project
62 rdf:type schema:Organization
63 Ne5d050690b2e4680bb5437035e30ec08 rdf:first sg:person.011304472045.51
64 rdf:rest rdf:nil
65 anzsrc-for:01 schema:inDefinedTermSet anzsrc-for:
66 schema:name Mathematical Sciences
67 rdf:type schema:DefinedTerm
68 anzsrc-for:0103 schema:inDefinedTermSet anzsrc-for:
69 schema:name Numerical and Computational Mathematics
70 rdf:type schema:DefinedTerm
71 sg:journal.1047630 schema:issn 0025-5610
72 1436-4646
73 schema:name Mathematical Programming
74 rdf:type schema:Periodical
75 sg:person.010340346467.73 schema:affiliation https://www.grid.ac/institutes/grid.495456.f
76 schema:familyName Cox
77 schema:givenName Bruce
78 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.010340346467.73
79 rdf:type schema:Person
80 sg:person.011304472045.51 schema:affiliation https://www.grid.ac/institutes/grid.213917.f
81 schema:familyName Nemirovski
82 schema:givenName Arkadi
83 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.011304472045.51
84 rdf:type schema:Person
85 sg:person.016013144233.55 schema:affiliation https://www.grid.ac/institutes/grid.464181.e
86 schema:familyName Juditsky
87 schema:givenName Anatoli
88 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016013144233.55
89 rdf:type schema:Person
90 sg:pub.10.1007/978-1-4419-8853-9 schema:sameAs https://app.dimensions.ai/details/publication/pub.1049823794
91 https://doi.org/10.1007/978-1-4419-8853-9
92 rdf:type schema:CreativeWork
93 sg:pub.10.1007/bf01585555 schema:sameAs https://app.dimensions.ai/details/publication/pub.1038375449
94 https://doi.org/10.1007/bf01585555
95 rdf:type schema:CreativeWork
96 sg:pub.10.1007/s10107-004-0552-5 schema:sameAs https://app.dimensions.ai/details/publication/pub.1030642071
97 https://doi.org/10.1007/s10107-004-0552-5
98 rdf:type schema:CreativeWork
99 sg:pub.10.1007/s10107-004-0553-4 schema:sameAs https://app.dimensions.ai/details/publication/pub.1021331750
100 https://doi.org/10.1007/s10107-004-0553-4
101 rdf:type schema:CreativeWork
102 sg:pub.10.1007/s10107-006-0034-z schema:sameAs https://app.dimensions.ai/details/publication/pub.1042069345
103 https://doi.org/10.1007/s10107-006-0034-z
104 rdf:type schema:CreativeWork
105 sg:pub.10.1007/s10107-007-0149-x schema:sameAs https://app.dimensions.ai/details/publication/pub.1016713895
106 https://doi.org/10.1007/s10107-007-0149-x
107 rdf:type schema:CreativeWork
108 sg:pub.10.1007/s11228-010-0147-7 schema:sameAs https://app.dimensions.ai/details/publication/pub.1032025040
109 https://doi.org/10.1007/s11228-010-0147-7
110 rdf:type schema:CreativeWork
111 https://doi.org/10.1002/nav.3800030109 schema:sameAs https://app.dimensions.ai/details/publication/pub.1047094257
112 rdf:type schema:CreativeWork
113 https://doi.org/10.1016/0022-247x(78)90137-3 schema:sameAs https://app.dimensions.ai/details/publication/pub.1009751684
114 rdf:type schema:CreativeWork
115 https://doi.org/10.1016/0041-5553(67)90040-7 schema:sameAs https://app.dimensions.ai/details/publication/pub.1010322999
116 rdf:type schema:CreativeWork
117 https://doi.org/10.1017/s096249291300007x schema:sameAs https://app.dimensions.ai/details/publication/pub.1054905297
118 rdf:type schema:CreativeWork
119 https://doi.org/10.1137/0803026 schema:sameAs https://app.dimensions.ai/details/publication/pub.1062854211
120 rdf:type schema:CreativeWork
121 https://doi.org/10.1137/s1052623403425629 schema:sameAs https://app.dimensions.ai/details/publication/pub.1062883356
122 rdf:type schema:CreativeWork
123 https://doi.org/10.1145/1273496.1273499 schema:sameAs https://app.dimensions.ai/details/publication/pub.1031215462
124 rdf:type schema:CreativeWork
125 https://doi.org/10.1214/11-aos944 schema:sameAs https://app.dimensions.ai/details/publication/pub.1002363869
126 rdf:type schema:CreativeWork
127 https://doi.org/10.1287/moor.1090.0427 schema:sameAs https://app.dimensions.ai/details/publication/pub.1064723060
128 rdf:type schema:CreativeWork
129 https://doi.org/10.24033/bsmf.1625 schema:sameAs https://app.dimensions.ai/details/publication/pub.1083660582
130 rdf:type schema:CreativeWork
131 https://www.grid.ac/institutes/grid.213917.f schema:alternateName Georgia Institute of Technology
132 schema:name Georgia Institute of Technology, 30332, Atlanta, GA, USA
133 rdf:type schema:Organization
134 https://www.grid.ac/institutes/grid.464181.e schema:alternateName Laboratoire Jean Kuntzmann
135 schema:name LJK, Université J. Fourier, B.P. 53, 38041, Grenoble Cedex 9, France
136 rdf:type schema:Organization
137 https://www.grid.ac/institutes/grid.495456.f schema:alternateName United States Department of the Air Force
138 schema:name US Air Force, Washington, DC, USA
139 rdf:type schema:Organization
 




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


...