Inclusive GAN: Improving Data and Minority Coverage in Generative Models View Full Text


Ontology type: schema:Chapter      Open Access: True


Chapter Info

DATE

2020-11-17

AUTHORS

Ning Yu , Ke Li , Peng Zhou , Jitendra Malik , Larry Davis , Mario Fritz

ABSTRACT

Generative Adversarial Networks (GANs) have brought about rapid progress towards generating photorealistic images. Yet the equitable allocation of their modeling capacity among subgroups has received less attention, which could lead to potential biases against underrepresented minorities if left uncontrolled. In this work, we first formalize the problem of minority inclusion as one of data coverage, and then propose to improve data coverage by harmonizing adversarial training with reconstructive generation. The experiments show that our method outperforms the existing state-of-the-art methods in terms of data coverage on both seen and unseen data. We develop an extension that allows explicit control over the minority subgroups that the model should ensure to include, and validate its effectiveness at little compromise from the overall performance on the entire dataset. Code, models, and supplemental videos are available at GitHub. More... »

PAGES

377-393

Identifiers

URI

http://scigraph.springernature.com/pub.10.1007/978-3-030-58542-6_23

DOI

http://dx.doi.org/10.1007/978-3-030-58542-6_23

DIMENSIONS

https://app.dimensions.ai/details/publication/pub.1132659589


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/2008/01", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Mathematical Sciences", 
        "type": "DefinedTerm"
      }, 
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/0104", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Statistics", 
        "type": "DefinedTerm"
      }
    ], 
    "author": [
      {
        "affiliation": {
          "alternateName": "Max Planck Institute for Informatics, Saarbr\u00fccken, Germany", 
          "id": "http://www.grid.ac/institutes/grid.419528.3", 
          "name": [
            "University of Maryland, College Park, USA", 
            "Max Planck Institute for Informatics, Saarbr\u00fccken, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Yu", 
        "givenName": "Ning", 
        "id": "sg:person.016547261025.90", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016547261025.90"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Google, Seattle, USA", 
          "id": "http://www.grid.ac/institutes/grid.420451.6", 
          "name": [
            "University of California, Berkeley, USA", 
            "Institute for Advanced Study, Princeton, USA", 
            "Google, Seattle, USA"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Li", 
        "givenName": "Ke", 
        "id": "sg:person.016022530322.54", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016022530322.54"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "University of Maryland, College Park, USA", 
          "id": "http://www.grid.ac/institutes/grid.164295.d", 
          "name": [
            "University of Maryland, College Park, USA"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Zhou", 
        "givenName": "Peng", 
        "id": "sg:person.010172245763.59", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.010172245763.59"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "University of California, Berkeley, USA", 
          "id": "http://www.grid.ac/institutes/grid.47840.3f", 
          "name": [
            "University of California, Berkeley, USA"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Malik", 
        "givenName": "Jitendra", 
        "id": "sg:person.01364521761.84", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01364521761.84"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "University of Maryland, College Park, USA", 
          "id": "http://www.grid.ac/institutes/grid.164295.d", 
          "name": [
            "University of Maryland, College Park, USA"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Davis", 
        "givenName": "Larry", 
        "id": "sg:person.07603773173.63", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.07603773173.63"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "CISPA Helmholtz Center for Information Security, Saarbr\u00fccken, Germany", 
          "id": "http://www.grid.ac/institutes/grid.507511.7", 
          "name": [
            "CISPA Helmholtz Center for Information Security, Saarbr\u00fccken, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Fritz", 
        "givenName": "Mario", 
        "id": "sg:person.013361072755.17", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013361072755.17"
        ], 
        "type": "Person"
      }
    ], 
    "datePublished": "2020-11-17", 
    "datePublishedReg": "2020-11-17", 
    "description": "Generative Adversarial Networks (GANs) have brought about rapid progress towards generating photorealistic images. Yet the equitable allocation of their modeling capacity among subgroups has received less attention, which could lead to potential biases against underrepresented minorities if left uncontrolled. In this work, we first formalize the problem of minority inclusion as one of data coverage, and then propose to improve data coverage by harmonizing adversarial training with reconstructive generation. The experiments show that our method outperforms the existing state-of-the-art methods in terms of data coverage on both seen and unseen data. We develop an extension that allows explicit control over the minority subgroups that the model should ensure to include, and validate its effectiveness at little compromise from the overall performance on the entire dataset. Code, models, and supplemental videos are available at GitHub.", 
    "editor": [
      {
        "familyName": "Vedaldi", 
        "givenName": "Andrea", 
        "type": "Person"
      }, 
      {
        "familyName": "Bischof", 
        "givenName": "Horst", 
        "type": "Person"
      }, 
      {
        "familyName": "Brox", 
        "givenName": "Thomas", 
        "type": "Person"
      }, 
      {
        "familyName": "Frahm", 
        "givenName": "Jan-Michael", 
        "type": "Person"
      }
    ], 
    "genre": "chapter", 
    "id": "sg:pub.10.1007/978-3-030-58542-6_23", 
    "isAccessibleForFree": true, 
    "isPartOf": {
      "isbn": [
        "978-3-030-58541-9", 
        "978-3-030-58542-6"
      ], 
      "name": "Computer Vision \u2013 ECCV 2020", 
      "type": "Book"
    }, 
    "keywords": [
      "modeling capacity", 
      "data coverage", 
      "explicit control", 
      "generative model", 
      "generative adversarial network", 
      "art methods", 
      "unseen data", 
      "Improving Data", 
      "model", 
      "little compromise", 
      "overall performance", 
      "entire dataset", 
      "problem", 
      "adversarial training", 
      "extension", 
      "adversarial network", 
      "network", 
      "terms", 
      "code", 
      "allocation", 
      "rapid progress", 
      "GitHub", 
      "photorealistic images", 
      "effectiveness", 
      "state", 
      "performance", 
      "data", 
      "compromise", 
      "dataset", 
      "work", 
      "experiments", 
      "coverage", 
      "control", 
      "generation", 
      "inclusion", 
      "biases", 
      "progress", 
      "potential biases", 
      "subgroups", 
      "less attention", 
      "images", 
      "equitable allocation", 
      "attention", 
      "supplemental video", 
      "video", 
      "capacity", 
      "training", 
      "minority subgroups", 
      "minority", 
      "method", 
      "underrepresented minorities", 
      "minority inclusion"
    ], 
    "name": "Inclusive GAN: Improving Data and Minority Coverage in Generative Models", 
    "pagination": "377-393", 
    "productId": [
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "pub.1132659589"
        ]
      }, 
      {
        "name": "doi", 
        "type": "PropertyValue", 
        "value": [
          "10.1007/978-3-030-58542-6_23"
        ]
      }
    ], 
    "publisher": {
      "name": "Springer Nature", 
      "type": "Organisation"
    }, 
    "sameAs": [
      "https://doi.org/10.1007/978-3-030-58542-6_23", 
      "https://app.dimensions.ai/details/publication/pub.1132659589"
    ], 
    "sdDataset": "chapters", 
    "sdDatePublished": "2022-12-01T06:53", 
    "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
    "sdPublisher": {
      "name": "Springer Nature - SN SciGraph project", 
      "type": "Organization"
    }, 
    "sdSource": "s3://com-springernature-scigraph/baseset/20221201/entities/gbq_results/chapter/chapter_397.jsonl", 
    "type": "Chapter", 
    "url": "https://doi.org/10.1007/978-3-030-58542-6_23"
  }
]
 

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.10.1007/978-3-030-58542-6_23'

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

curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1007/978-3-030-58542-6_23'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/978-3-030-58542-6_23'

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

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/978-3-030-58542-6_23'


 

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

176 TRIPLES      22 PREDICATES      76 URIs      69 LITERALS      7 BLANK NODES

Subject Predicate Object
1 sg:pub.10.1007/978-3-030-58542-6_23 schema:about anzsrc-for:01
2 anzsrc-for:0104
3 schema:author N7e425284a4934c6aaf4288084e7724c0
4 schema:datePublished 2020-11-17
5 schema:datePublishedReg 2020-11-17
6 schema:description Generative Adversarial Networks (GANs) have brought about rapid progress towards generating photorealistic images. Yet the equitable allocation of their modeling capacity among subgroups has received less attention, which could lead to potential biases against underrepresented minorities if left uncontrolled. In this work, we first formalize the problem of minority inclusion as one of data coverage, and then propose to improve data coverage by harmonizing adversarial training with reconstructive generation. The experiments show that our method outperforms the existing state-of-the-art methods in terms of data coverage on both seen and unseen data. We develop an extension that allows explicit control over the minority subgroups that the model should ensure to include, and validate its effectiveness at little compromise from the overall performance on the entire dataset. Code, models, and supplemental videos are available at GitHub.
7 schema:editor N3fd713daa7b043fea12a21c955d79ed6
8 schema:genre chapter
9 schema:isAccessibleForFree true
10 schema:isPartOf Nc90cefed14fc4b4291917676cc895fc3
11 schema:keywords GitHub
12 Improving Data
13 adversarial network
14 adversarial training
15 allocation
16 art methods
17 attention
18 biases
19 capacity
20 code
21 compromise
22 control
23 coverage
24 data
25 data coverage
26 dataset
27 effectiveness
28 entire dataset
29 equitable allocation
30 experiments
31 explicit control
32 extension
33 generation
34 generative adversarial network
35 generative model
36 images
37 inclusion
38 less attention
39 little compromise
40 method
41 minority
42 minority inclusion
43 minority subgroups
44 model
45 modeling capacity
46 network
47 overall performance
48 performance
49 photorealistic images
50 potential biases
51 problem
52 progress
53 rapid progress
54 state
55 subgroups
56 supplemental video
57 terms
58 training
59 underrepresented minorities
60 unseen data
61 video
62 work
63 schema:name Inclusive GAN: Improving Data and Minority Coverage in Generative Models
64 schema:pagination 377-393
65 schema:productId N12c671d9219540c8bf775030e5d70302
66 Na053c09baa394727bd05f083338de4ad
67 schema:publisher Nb054597202c5401e8b1084b9736dda7b
68 schema:sameAs https://app.dimensions.ai/details/publication/pub.1132659589
69 https://doi.org/10.1007/978-3-030-58542-6_23
70 schema:sdDatePublished 2022-12-01T06:53
71 schema:sdLicense https://scigraph.springernature.com/explorer/license/
72 schema:sdPublisher Ndd85404fe54547deaec6dec24887a582
73 schema:url https://doi.org/10.1007/978-3-030-58542-6_23
74 sgo:license sg:explorer/license/
75 sgo:sdDataset chapters
76 rdf:type schema:Chapter
77 N12c671d9219540c8bf775030e5d70302 schema:name dimensions_id
78 schema:value pub.1132659589
79 rdf:type schema:PropertyValue
80 N144fea4142f346d9bee9407755b85e16 rdf:first sg:person.07603773173.63
81 rdf:rest N82f5224e915e4c4f889d9fa2455d6316
82 N2dc7fc3179aa4367853d089be63c8eb5 rdf:first N542beaf95aea44e88ddf002674d93e12
83 rdf:rest N8930868f50db4ebc812fec3660c4589a
84 N3fd713daa7b043fea12a21c955d79ed6 rdf:first Nfc054382ff534bee9984c4355f0dd23c
85 rdf:rest N2dc7fc3179aa4367853d089be63c8eb5
86 N477c3b3b47c641f7a1eea1017976ff25 rdf:first sg:person.010172245763.59
87 rdf:rest N96f11237a5b74e3fb8b1fadeb0284c37
88 N48d0a737139c47649ede73104a25f417 rdf:first sg:person.016022530322.54
89 rdf:rest N477c3b3b47c641f7a1eea1017976ff25
90 N542beaf95aea44e88ddf002674d93e12 schema:familyName Bischof
91 schema:givenName Horst
92 rdf:type schema:Person
93 N7e425284a4934c6aaf4288084e7724c0 rdf:first sg:person.016547261025.90
94 rdf:rest N48d0a737139c47649ede73104a25f417
95 N80afd436d9624e3b883a8401a4cfb196 schema:familyName Brox
96 schema:givenName Thomas
97 rdf:type schema:Person
98 N82f5224e915e4c4f889d9fa2455d6316 rdf:first sg:person.013361072755.17
99 rdf:rest rdf:nil
100 N8930868f50db4ebc812fec3660c4589a rdf:first N80afd436d9624e3b883a8401a4cfb196
101 rdf:rest N926ac1a0a01d4543838d5632033aa640
102 N926ac1a0a01d4543838d5632033aa640 rdf:first Ndbe00136926546e99b3a1efb8a770b0c
103 rdf:rest rdf:nil
104 N96f11237a5b74e3fb8b1fadeb0284c37 rdf:first sg:person.01364521761.84
105 rdf:rest N144fea4142f346d9bee9407755b85e16
106 Na053c09baa394727bd05f083338de4ad schema:name doi
107 schema:value 10.1007/978-3-030-58542-6_23
108 rdf:type schema:PropertyValue
109 Nb054597202c5401e8b1084b9736dda7b schema:name Springer Nature
110 rdf:type schema:Organisation
111 Nc90cefed14fc4b4291917676cc895fc3 schema:isbn 978-3-030-58541-9
112 978-3-030-58542-6
113 schema:name Computer Vision – ECCV 2020
114 rdf:type schema:Book
115 Ndbe00136926546e99b3a1efb8a770b0c schema:familyName Frahm
116 schema:givenName Jan-Michael
117 rdf:type schema:Person
118 Ndd85404fe54547deaec6dec24887a582 schema:name Springer Nature - SN SciGraph project
119 rdf:type schema:Organization
120 Nfc054382ff534bee9984c4355f0dd23c schema:familyName Vedaldi
121 schema:givenName Andrea
122 rdf:type schema:Person
123 anzsrc-for:01 schema:inDefinedTermSet anzsrc-for:
124 schema:name Mathematical Sciences
125 rdf:type schema:DefinedTerm
126 anzsrc-for:0104 schema:inDefinedTermSet anzsrc-for:
127 schema:name Statistics
128 rdf:type schema:DefinedTerm
129 sg:person.010172245763.59 schema:affiliation grid-institutes:grid.164295.d
130 schema:familyName Zhou
131 schema:givenName Peng
132 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.010172245763.59
133 rdf:type schema:Person
134 sg:person.013361072755.17 schema:affiliation grid-institutes:grid.507511.7
135 schema:familyName Fritz
136 schema:givenName Mario
137 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013361072755.17
138 rdf:type schema:Person
139 sg:person.01364521761.84 schema:affiliation grid-institutes:grid.47840.3f
140 schema:familyName Malik
141 schema:givenName Jitendra
142 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01364521761.84
143 rdf:type schema:Person
144 sg:person.016022530322.54 schema:affiliation grid-institutes:grid.420451.6
145 schema:familyName Li
146 schema:givenName Ke
147 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016022530322.54
148 rdf:type schema:Person
149 sg:person.016547261025.90 schema:affiliation grid-institutes:grid.419528.3
150 schema:familyName Yu
151 schema:givenName Ning
152 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016547261025.90
153 rdf:type schema:Person
154 sg:person.07603773173.63 schema:affiliation grid-institutes:grid.164295.d
155 schema:familyName Davis
156 schema:givenName Larry
157 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.07603773173.63
158 rdf:type schema:Person
159 grid-institutes:grid.164295.d schema:alternateName University of Maryland, College Park, USA
160 schema:name University of Maryland, College Park, USA
161 rdf:type schema:Organization
162 grid-institutes:grid.419528.3 schema:alternateName Max Planck Institute for Informatics, Saarbrücken, Germany
163 schema:name Max Planck Institute for Informatics, Saarbrücken, Germany
164 University of Maryland, College Park, USA
165 rdf:type schema:Organization
166 grid-institutes:grid.420451.6 schema:alternateName Google, Seattle, USA
167 schema:name Google, Seattle, USA
168 Institute for Advanced Study, Princeton, USA
169 University of California, Berkeley, USA
170 rdf:type schema:Organization
171 grid-institutes:grid.47840.3f schema:alternateName University of California, Berkeley, USA
172 schema:name University of California, Berkeley, USA
173 rdf:type schema:Organization
174 grid-institutes:grid.507511.7 schema:alternateName CISPA Helmholtz Center for Information Security, Saarbrücken, Germany
175 schema:name CISPA Helmholtz Center for Information Security, Saarbrücken, Germany
176 rdf:type schema:Organization
 




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


...