Beyond support in two-stage variable selection View Full Text


Ontology type: schema:ScholarlyArticle      Open Access: True


Article Info

DATE

2015-11-20

AUTHORS

Jean-Michel Bécu, Yves Grandvalet, Christophe Ambroise, Cyril Dalmasso

ABSTRACT

Numerous variable selection methods rely on a two-stage procedure, where a sparsity-inducing penalty is used in the first stage to predict the support, which is then conveyed to the second stage for estimation or inference purposes. In this framework, the first stage screens variables to find a set of possibly relevant variables and the second stage operates on this set of candidate variables, to improve estimation accuracy or to assess the uncertainty associated to the selection of variables. We advocate that more information can be conveyed from the first stage to the second one: we use the magnitude of the coefficients estimated in the first stage to define an adaptive penalty that is applied at the second stage. We give the example of an inference procedure that highly benefits from the proposed transfer of information. The procedure is precisely analyzed in a simple setting, and our large-scale experiments empirically demonstrate that actual benefits can be expected in much more general situations, with sensitivity gains ranging from 50 to 100 % compared to state-of-the-art. More... »

PAGES

169-179

Identifiers

URI

http://scigraph.springernature.com/pub.10.1007/s11222-015-9614-1

DOI

http://dx.doi.org/10.1007/s11222-015-9614-1

DIMENSIONS

https://app.dimensions.ai/details/publication/pub.1004231488


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/2008/08", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Information and Computing Sciences", 
        "type": "DefinedTerm"
      }, 
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/0801", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Artificial Intelligence and Image Processing", 
        "type": "DefinedTerm"
      }
    ], 
    "author": [
      {
        "affiliation": {
          "alternateName": "Sorbonne universit\u00e9s, Universit\u00e9 de technologie de Compi\u00e8gne, CNRS, Heudiasyc UMR 7253, CS 60 319, 60 203, Compi\u00e8gne Cedex, France", 
          "id": "http://www.grid.ac/institutes/None", 
          "name": [
            "Sorbonne universit\u00e9s, Universit\u00e9 de technologie de Compi\u00e8gne, CNRS, Heudiasyc UMR 7253, CS 60 319, 60 203, Compi\u00e8gne Cedex, France"
          ], 
          "type": "Organization"
        }, 
        "familyName": "B\u00e9cu", 
        "givenName": "Jean-Michel", 
        "id": "sg:person.07561010270.53", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.07561010270.53"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Sorbonne universit\u00e9s, Universit\u00e9 de technologie de Compi\u00e8gne, CNRS, Heudiasyc UMR 7253, CS 60 319, 60 203, Compi\u00e8gne Cedex, France", 
          "id": "http://www.grid.ac/institutes/None", 
          "name": [
            "Sorbonne universit\u00e9s, Universit\u00e9 de technologie de Compi\u00e8gne, CNRS, Heudiasyc UMR 7253, CS 60 319, 60 203, Compi\u00e8gne Cedex, France"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Grandvalet", 
        "givenName": "Yves", 
        "id": "sg:person.015255215731.52", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.015255215731.52"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "LaMME, Universit\u00e9 d\u2019\u00c9vry val d\u2019Essonne, 23 Boulevard de France, 91000, \u00c9vry, France", 
          "id": "http://www.grid.ac/institutes/grid.8390.2", 
          "name": [
            "LaMME, Universit\u00e9 d\u2019\u00c9vry val d\u2019Essonne, 23 Boulevard de France, 91000, \u00c9vry, France"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Ambroise", 
        "givenName": "Christophe", 
        "id": "sg:person.016650156731.69", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016650156731.69"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "LaMME, Universit\u00e9 d\u2019\u00c9vry val d\u2019Essonne, 23 Boulevard de France, 91000, \u00c9vry, France", 
          "id": "http://www.grid.ac/institutes/grid.8390.2", 
          "name": [
            "LaMME, Universit\u00e9 d\u2019\u00c9vry val d\u2019Essonne, 23 Boulevard de France, 91000, \u00c9vry, France"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Dalmasso", 
        "givenName": "Cyril", 
        "id": "sg:person.01247430571.45", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01247430571.45"
        ], 
        "type": "Person"
      }
    ], 
    "citation": [
      {
        "id": "sg:pub.10.1007/978-0-387-49317-6", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1029469077", 
          "https://doi.org/10.1007/978-0-387-49317-6"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1038/nrg1916", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1029660577", 
          "https://doi.org/10.1038/nrg1916"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1186/1471-2105-12-372", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1050488595", 
          "https://doi.org/10.1186/1471-2105-12-372"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/978-1-4471-1599-1_27", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1053535549", 
          "https://doi.org/10.1007/978-1-4471-1599-1_27"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1186/gb-2010-11-10-r106", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1031289083", 
          "https://doi.org/10.1186/gb-2010-11-10-r106"
        ], 
        "type": "CreativeWork"
      }
    ], 
    "datePublished": "2015-11-20", 
    "datePublishedReg": "2015-11-20", 
    "description": "Numerous variable selection methods rely on a two-stage procedure, where a sparsity-inducing penalty is used in the first stage to predict the support, which is then conveyed to the second stage for estimation or inference purposes. In this framework, the first stage screens variables to find a set of possibly relevant variables and the second stage operates on this set of candidate variables, to improve estimation accuracy or to assess the uncertainty associated to the selection of variables. We advocate that more information can be conveyed from the first stage to the second one: we use the magnitude of the coefficients estimated in the first stage to define an adaptive penalty that is applied at the second stage. We give the example of an inference procedure that highly benefits from the proposed transfer of information. The procedure is precisely analyzed in a simple setting, and our large-scale experiments empirically demonstrate that actual benefits can be expected in much more general situations, with sensitivity gains ranging from 50 to 100\u00a0% compared to state-of-the-art.", 
    "genre": "article", 
    "id": "sg:pub.10.1007/s11222-015-9614-1", 
    "inLanguage": "en", 
    "isAccessibleForFree": true, 
    "isPartOf": [
      {
        "id": "sg:journal.1327447", 
        "issn": [
          "0960-3174", 
          "1573-1375"
        ], 
        "name": "Statistics and Computing", 
        "publisher": "Springer Nature", 
        "type": "Periodical"
      }, 
      {
        "issueNumber": "1", 
        "type": "PublicationIssue"
      }, 
      {
        "type": "PublicationVolume", 
        "volumeNumber": "27"
      }
    ], 
    "keywords": [
      "two-stage variable selection", 
      "variable selection methods", 
      "sparsity-inducing penalty", 
      "adaptive penalty", 
      "large-scale experiments", 
      "selection method", 
      "inference purposes", 
      "estimation accuracy", 
      "first stage", 
      "variable selection", 
      "second stage", 
      "transfer of information", 
      "inference procedures", 
      "simple setting", 
      "selection of variables", 
      "second one", 
      "information", 
      "set", 
      "more information", 
      "actual benefits", 
      "penalty", 
      "framework", 
      "more general situations", 
      "selection", 
      "accuracy", 
      "candidate variables", 
      "support", 
      "estimation", 
      "general situation", 
      "art", 
      "benefits", 
      "example", 
      "situation", 
      "uncertainty", 
      "two-stage procedure", 
      "method", 
      "relevant variables", 
      "experiments", 
      "one", 
      "procedure", 
      "stage", 
      "gain", 
      "variables", 
      "purpose", 
      "state", 
      "setting", 
      "coefficient", 
      "sensitivity gain", 
      "transfer", 
      "magnitude", 
      "Numerous variable selection methods"
    ], 
    "name": "Beyond support in two-stage variable selection", 
    "pagination": "169-179", 
    "productId": [
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "pub.1004231488"
        ]
      }, 
      {
        "name": "doi", 
        "type": "PropertyValue", 
        "value": [
          "10.1007/s11222-015-9614-1"
        ]
      }
    ], 
    "sameAs": [
      "https://doi.org/10.1007/s11222-015-9614-1", 
      "https://app.dimensions.ai/details/publication/pub.1004231488"
    ], 
    "sdDataset": "articles", 
    "sdDatePublished": "2022-01-01T18:35", 
    "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
    "sdPublisher": {
      "name": "Springer Nature - SN SciGraph project", 
      "type": "Organization"
    }, 
    "sdSource": "s3://com-springernature-scigraph/baseset/20220101/entities/gbq_results/article/article_651.jsonl", 
    "type": "ScholarlyArticle", 
    "url": "https://doi.org/10.1007/s11222-015-9614-1"
  }
]
 

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/s11222-015-9614-1'

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/s11222-015-9614-1'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/s11222-015-9614-1'

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

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/s11222-015-9614-1'


 

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

153 TRIPLES      22 PREDICATES      81 URIs      68 LITERALS      6 BLANK NODES

Subject Predicate Object
1 sg:pub.10.1007/s11222-015-9614-1 schema:about anzsrc-for:08
2 anzsrc-for:0801
3 schema:author N5c428bfbbb5749ef9e3d52fbce05f879
4 schema:citation sg:pub.10.1007/978-0-387-49317-6
5 sg:pub.10.1007/978-1-4471-1599-1_27
6 sg:pub.10.1038/nrg1916
7 sg:pub.10.1186/1471-2105-12-372
8 sg:pub.10.1186/gb-2010-11-10-r106
9 schema:datePublished 2015-11-20
10 schema:datePublishedReg 2015-11-20
11 schema:description Numerous variable selection methods rely on a two-stage procedure, where a sparsity-inducing penalty is used in the first stage to predict the support, which is then conveyed to the second stage for estimation or inference purposes. In this framework, the first stage screens variables to find a set of possibly relevant variables and the second stage operates on this set of candidate variables, to improve estimation accuracy or to assess the uncertainty associated to the selection of variables. We advocate that more information can be conveyed from the first stage to the second one: we use the magnitude of the coefficients estimated in the first stage to define an adaptive penalty that is applied at the second stage. We give the example of an inference procedure that highly benefits from the proposed transfer of information. The procedure is precisely analyzed in a simple setting, and our large-scale experiments empirically demonstrate that actual benefits can be expected in much more general situations, with sensitivity gains ranging from 50 to 100 % compared to state-of-the-art.
12 schema:genre article
13 schema:inLanguage en
14 schema:isAccessibleForFree true
15 schema:isPartOf N7a9d09e91e834a63b651dd9243fcdcc8
16 Ned4cf3c01c344727bafd1a2ff2f54867
17 sg:journal.1327447
18 schema:keywords Numerous variable selection methods
19 accuracy
20 actual benefits
21 adaptive penalty
22 art
23 benefits
24 candidate variables
25 coefficient
26 estimation
27 estimation accuracy
28 example
29 experiments
30 first stage
31 framework
32 gain
33 general situation
34 inference procedures
35 inference purposes
36 information
37 large-scale experiments
38 magnitude
39 method
40 more general situations
41 more information
42 one
43 penalty
44 procedure
45 purpose
46 relevant variables
47 second one
48 second stage
49 selection
50 selection method
51 selection of variables
52 sensitivity gain
53 set
54 setting
55 simple setting
56 situation
57 sparsity-inducing penalty
58 stage
59 state
60 support
61 transfer
62 transfer of information
63 two-stage procedure
64 two-stage variable selection
65 uncertainty
66 variable selection
67 variable selection methods
68 variables
69 schema:name Beyond support in two-stage variable selection
70 schema:pagination 169-179
71 schema:productId N80bb4b57f62e493d83fd331182656909
72 N92542e56a957420ca9738e3ed43c729c
73 schema:sameAs https://app.dimensions.ai/details/publication/pub.1004231488
74 https://doi.org/10.1007/s11222-015-9614-1
75 schema:sdDatePublished 2022-01-01T18:35
76 schema:sdLicense https://scigraph.springernature.com/explorer/license/
77 schema:sdPublisher N0f5adfb80abe4fe59e42e37155124e18
78 schema:url https://doi.org/10.1007/s11222-015-9614-1
79 sgo:license sg:explorer/license/
80 sgo:sdDataset articles
81 rdf:type schema:ScholarlyArticle
82 N0019b835cef4418c9d5ddc356c31bf79 rdf:first sg:person.015255215731.52
83 rdf:rest N2c893ea518fb437c829973687c95452d
84 N0f5adfb80abe4fe59e42e37155124e18 schema:name Springer Nature - SN SciGraph project
85 rdf:type schema:Organization
86 N2bed320f89f94feebedec5cb2dbb8755 rdf:first sg:person.01247430571.45
87 rdf:rest rdf:nil
88 N2c893ea518fb437c829973687c95452d rdf:first sg:person.016650156731.69
89 rdf:rest N2bed320f89f94feebedec5cb2dbb8755
90 N5c428bfbbb5749ef9e3d52fbce05f879 rdf:first sg:person.07561010270.53
91 rdf:rest N0019b835cef4418c9d5ddc356c31bf79
92 N7a9d09e91e834a63b651dd9243fcdcc8 schema:volumeNumber 27
93 rdf:type schema:PublicationVolume
94 N80bb4b57f62e493d83fd331182656909 schema:name doi
95 schema:value 10.1007/s11222-015-9614-1
96 rdf:type schema:PropertyValue
97 N92542e56a957420ca9738e3ed43c729c schema:name dimensions_id
98 schema:value pub.1004231488
99 rdf:type schema:PropertyValue
100 Ned4cf3c01c344727bafd1a2ff2f54867 schema:issueNumber 1
101 rdf:type schema:PublicationIssue
102 anzsrc-for:08 schema:inDefinedTermSet anzsrc-for:
103 schema:name Information and Computing Sciences
104 rdf:type schema:DefinedTerm
105 anzsrc-for:0801 schema:inDefinedTermSet anzsrc-for:
106 schema:name Artificial Intelligence and Image Processing
107 rdf:type schema:DefinedTerm
108 sg:journal.1327447 schema:issn 0960-3174
109 1573-1375
110 schema:name Statistics and Computing
111 schema:publisher Springer Nature
112 rdf:type schema:Periodical
113 sg:person.01247430571.45 schema:affiliation grid-institutes:grid.8390.2
114 schema:familyName Dalmasso
115 schema:givenName Cyril
116 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01247430571.45
117 rdf:type schema:Person
118 sg:person.015255215731.52 schema:affiliation grid-institutes:None
119 schema:familyName Grandvalet
120 schema:givenName Yves
121 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.015255215731.52
122 rdf:type schema:Person
123 sg:person.016650156731.69 schema:affiliation grid-institutes:grid.8390.2
124 schema:familyName Ambroise
125 schema:givenName Christophe
126 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016650156731.69
127 rdf:type schema:Person
128 sg:person.07561010270.53 schema:affiliation grid-institutes:None
129 schema:familyName Bécu
130 schema:givenName Jean-Michel
131 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.07561010270.53
132 rdf:type schema:Person
133 sg:pub.10.1007/978-0-387-49317-6 schema:sameAs https://app.dimensions.ai/details/publication/pub.1029469077
134 https://doi.org/10.1007/978-0-387-49317-6
135 rdf:type schema:CreativeWork
136 sg:pub.10.1007/978-1-4471-1599-1_27 schema:sameAs https://app.dimensions.ai/details/publication/pub.1053535549
137 https://doi.org/10.1007/978-1-4471-1599-1_27
138 rdf:type schema:CreativeWork
139 sg:pub.10.1038/nrg1916 schema:sameAs https://app.dimensions.ai/details/publication/pub.1029660577
140 https://doi.org/10.1038/nrg1916
141 rdf:type schema:CreativeWork
142 sg:pub.10.1186/1471-2105-12-372 schema:sameAs https://app.dimensions.ai/details/publication/pub.1050488595
143 https://doi.org/10.1186/1471-2105-12-372
144 rdf:type schema:CreativeWork
145 sg:pub.10.1186/gb-2010-11-10-r106 schema:sameAs https://app.dimensions.ai/details/publication/pub.1031289083
146 https://doi.org/10.1186/gb-2010-11-10-r106
147 rdf:type schema:CreativeWork
148 grid-institutes:None schema:alternateName Sorbonne universités, Université de technologie de Compiègne, CNRS, Heudiasyc UMR 7253, CS 60 319, 60 203, Compiègne Cedex, France
149 schema:name Sorbonne universités, Université de technologie de Compiègne, CNRS, Heudiasyc UMR 7253, CS 60 319, 60 203, Compiègne Cedex, France
150 rdf:type schema:Organization
151 grid-institutes:grid.8390.2 schema:alternateName LaMME, Université d’Évry val d’Essonne, 23 Boulevard de France, 91000, Évry, France
152 schema:name LaMME, Université d’Évry val d’Essonne, 23 Boulevard de France, 91000, Évry, France
153 rdf:type schema:Organization
 




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


...