Using model uncertainty for robust optimization in approximate inference control View Full Text


Ontology type: schema:ScholarlyArticle     


Article Info

DATE

2017-09

AUTHORS

Hideaki Itoh, Yoshitaka Sakai, Toru Kadoya, Hisao Fukumoto, Hiroshi Wakuya, Tatsuya Furukawa

ABSTRACT

Recently, the optimization-by-inference approach has been proposed as a new means for solving high-dimensional optimization problems quickly. Approximate Inference COntrol (AICO) is one of the most successful and promising methods that implement the optimization-by-inference approach. AICO is able to solve stochastic optimal control problems and has already been successfully used in many applications. However, it is known that the iterative inference of AICO sometimes fails to converge to the optimal solution. To make the optimization more robust, in this paper, we propose to take model uncertainty into account. In AICO, the cost function to be minimized is accurate around a particular state of a given stochastic system, but the accuracy is uncertain in regions far from that state. Because using such an uncertain function is harmful to the convergence, we modify AICO, so that it does not use the function in uncertain regions. Our method is easy to implement and does not add much computational time to the original AICO. Experiments using two different scenarios show that our method substantially improves AICO in terms of the rate at which the algorithm produces convergent results. More... »

PAGES

327-335

References to SciGraph publications

Identifiers

URI

http://scigraph.springernature.com/pub.10.1007/s10015-017-0361-6

DOI

http://dx.doi.org/10.1007/s10015-017-0361-6

DIMENSIONS

https://app.dimensions.ai/details/publication/pub.1084023566


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": "Saga University", 
          "id": "https://www.grid.ac/institutes/grid.412339.e", 
          "name": [
            "Department of Electrical and Electronic Engineering, Graduate School of Science and Engineering, Saga University, 1 Honjo-machi, 840-8502, Saga, Japan"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Itoh", 
        "givenName": "Hideaki", 
        "id": "sg:person.013416235535.24", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013416235535.24"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Fujitsu (Japan)", 
          "id": "https://www.grid.ac/institutes/grid.418251.b", 
          "name": [
            "Fujitsu Limited, 1-5-2 Higashi-Shimbashi, Minato-ku, 105-7123, Tokyo, Japan"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Sakai", 
        "givenName": "Yoshitaka", 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Hitachi (Japan)", 
          "id": "https://www.grid.ac/institutes/grid.417547.4", 
          "name": [
            "Hitachi Systems, Ltd., 1-2-1 Osaki, Sinagawa-ku, 141-8672, Tokyo, Japan"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Kadoya", 
        "givenName": "Toru", 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Saga University", 
          "id": "https://www.grid.ac/institutes/grid.412339.e", 
          "name": [
            "Department of Electrical and Electronic Engineering, Graduate School of Science and Engineering, Saga University, 1 Honjo-machi, 840-8502, Saga, Japan"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Fukumoto", 
        "givenName": "Hisao", 
        "id": "sg:person.010430333535.31", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.010430333535.31"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Saga University", 
          "id": "https://www.grid.ac/institutes/grid.412339.e", 
          "name": [
            "Department of Electrical and Electronic Engineering, Graduate School of Science and Engineering, Saga University, 1 Honjo-machi, 840-8502, Saga, Japan"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Wakuya", 
        "givenName": "Hiroshi", 
        "id": "sg:person.012705031237.88", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.012705031237.88"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Saga University", 
          "id": "https://www.grid.ac/institutes/grid.412339.e", 
          "name": [
            "Department of Electrical and Electronic Engineering, Graduate School of Science and Engineering, Saga University, 1 Honjo-machi, 840-8502, Saga, Japan"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Furukawa", 
        "givenName": "Tatsuya", 
        "id": "sg:person.014746150256.43", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014746150256.43"
        ], 
        "type": "Person"
      }
    ], 
    "citation": [
      {
        "id": "sg:pub.10.1007/s10994-012-5278-7", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1000956677", 
          "https://doi.org/10.1007/s10994-012-5278-7"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/s10994-012-5278-7", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1000956677", 
          "https://doi.org/10.1007/s10994-012-5278-7"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/s10514-012-9315-y", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1017190414", 
          "https://doi.org/10.1007/s10514-012-9315-y"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1145/1553374.1553508", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1026898291"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1177/0278364913482017", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1027688393"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1177/0278364913482017", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1027688393"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1145/1143844.1143963", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1029214029"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/humanoids.2014.7041389", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1094167611"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.15607/rss.2012.viii.045", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1099108088"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.15607/rss.2012.viii.059", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1099108102"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1613/jair.4649", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1105689999"
        ], 
        "type": "CreativeWork"
      }
    ], 
    "datePublished": "2017-09", 
    "datePublishedReg": "2017-09-01", 
    "description": "Recently, the optimization-by-inference approach has been proposed as a new means for solving high-dimensional optimization problems quickly. Approximate Inference COntrol (AICO) is one of the most successful and promising methods that implement the optimization-by-inference approach. AICO is able to solve stochastic optimal control problems and has already been successfully used in many applications. However, it is known that the iterative inference of AICO sometimes fails to converge to the optimal solution. To make the optimization more robust, in this paper, we propose to take model uncertainty into account. In AICO, the cost function to be minimized is accurate around a particular state of a given stochastic system, but the accuracy is uncertain in regions far from that state. Because using such an uncertain function is harmful to the convergence, we modify AICO, so that it does not use the function in uncertain regions. Our method is easy to implement and does not add much computational time to the original AICO. Experiments using two different scenarios show that our method substantially improves AICO in terms of the rate at which the algorithm produces convergent results.", 
    "genre": "research_article", 
    "id": "sg:pub.10.1007/s10015-017-0361-6", 
    "inLanguage": [
      "en"
    ], 
    "isAccessibleForFree": false, 
    "isFundedItemOf": [
      {
        "id": "sg:grant.5865312", 
        "type": "MonetaryGrant"
      }
    ], 
    "isPartOf": [
      {
        "id": "sg:journal.1135365", 
        "issn": [
          "1433-5298", 
          "1614-7456"
        ], 
        "name": "Artificial Life and Robotics", 
        "type": "Periodical"
      }, 
      {
        "issueNumber": "3", 
        "type": "PublicationIssue"
      }, 
      {
        "type": "PublicationVolume", 
        "volumeNumber": "22"
      }
    ], 
    "name": "Using model uncertainty for robust optimization in approximate inference control", 
    "pagination": "327-335", 
    "productId": [
      {
        "name": "readcube_id", 
        "type": "PropertyValue", 
        "value": [
          "5719ba5bb39da7e30921fe897e9ad540131085e59efb7b010dd65316dbc5458e"
        ]
      }, 
      {
        "name": "doi", 
        "type": "PropertyValue", 
        "value": [
          "10.1007/s10015-017-0361-6"
        ]
      }, 
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "pub.1084023566"
        ]
      }
    ], 
    "sameAs": [
      "https://doi.org/10.1007/s10015-017-0361-6", 
      "https://app.dimensions.ai/details/publication/pub.1084023566"
    ], 
    "sdDataset": "articles", 
    "sdDatePublished": "2019-04-11T09:50", 
    "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/v90/0000000347_0000000347/records_89786_00000002.jsonl", 
    "type": "ScholarlyArticle", 
    "url": "https://link.springer.com/10.1007%2Fs10015-017-0361-6"
  }
]
 

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/s10015-017-0361-6'

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/s10015-017-0361-6'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/s10015-017-0361-6'

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

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/s10015-017-0361-6'


 

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

131 TRIPLES      21 PREDICATES      36 URIs      19 LITERALS      7 BLANK NODES

Subject Predicate Object
1 sg:pub.10.1007/s10015-017-0361-6 schema:about anzsrc-for:01
2 anzsrc-for:0103
3 schema:author N87512957a5f241aeb584cf10127ed2b7
4 schema:citation sg:pub.10.1007/s10514-012-9315-y
5 sg:pub.10.1007/s10994-012-5278-7
6 https://doi.org/10.1109/humanoids.2014.7041389
7 https://doi.org/10.1145/1143844.1143963
8 https://doi.org/10.1145/1553374.1553508
9 https://doi.org/10.1177/0278364913482017
10 https://doi.org/10.15607/rss.2012.viii.045
11 https://doi.org/10.15607/rss.2012.viii.059
12 https://doi.org/10.1613/jair.4649
13 schema:datePublished 2017-09
14 schema:datePublishedReg 2017-09-01
15 schema:description Recently, the optimization-by-inference approach has been proposed as a new means for solving high-dimensional optimization problems quickly. Approximate Inference COntrol (AICO) is one of the most successful and promising methods that implement the optimization-by-inference approach. AICO is able to solve stochastic optimal control problems and has already been successfully used in many applications. However, it is known that the iterative inference of AICO sometimes fails to converge to the optimal solution. To make the optimization more robust, in this paper, we propose to take model uncertainty into account. In AICO, the cost function to be minimized is accurate around a particular state of a given stochastic system, but the accuracy is uncertain in regions far from that state. Because using such an uncertain function is harmful to the convergence, we modify AICO, so that it does not use the function in uncertain regions. Our method is easy to implement and does not add much computational time to the original AICO. Experiments using two different scenarios show that our method substantially improves AICO in terms of the rate at which the algorithm produces convergent results.
16 schema:genre research_article
17 schema:inLanguage en
18 schema:isAccessibleForFree false
19 schema:isPartOf N84b286b15861488a957091bf97f8570c
20 Nae8e2e9a95b64f3b9708f3c91757a400
21 sg:journal.1135365
22 schema:name Using model uncertainty for robust optimization in approximate inference control
23 schema:pagination 327-335
24 schema:productId N30a35c86be784a018303ac44fe02cd9d
25 N95bb8b9006cb46d6b75b27c74b1a1d79
26 Nb637f1b5414b42aab9168c86c293f5f2
27 schema:sameAs https://app.dimensions.ai/details/publication/pub.1084023566
28 https://doi.org/10.1007/s10015-017-0361-6
29 schema:sdDatePublished 2019-04-11T09:50
30 schema:sdLicense https://scigraph.springernature.com/explorer/license/
31 schema:sdPublisher Nfe6d6ac0025944cbab5b57a273398e65
32 schema:url https://link.springer.com/10.1007%2Fs10015-017-0361-6
33 sgo:license sg:explorer/license/
34 sgo:sdDataset articles
35 rdf:type schema:ScholarlyArticle
36 N0b27a5527f254c6e877df244a9fdd724 rdf:first sg:person.010430333535.31
37 rdf:rest N4152d3267c15423b9647d9a25a6de4a0
38 N0c0b5ebd0b7c4493b8beb8274a11a06a rdf:first sg:person.014746150256.43
39 rdf:rest rdf:nil
40 N30a35c86be784a018303ac44fe02cd9d schema:name readcube_id
41 schema:value 5719ba5bb39da7e30921fe897e9ad540131085e59efb7b010dd65316dbc5458e
42 rdf:type schema:PropertyValue
43 N4152d3267c15423b9647d9a25a6de4a0 rdf:first sg:person.012705031237.88
44 rdf:rest N0c0b5ebd0b7c4493b8beb8274a11a06a
45 N84b286b15861488a957091bf97f8570c schema:issueNumber 3
46 rdf:type schema:PublicationIssue
47 N87512957a5f241aeb584cf10127ed2b7 rdf:first sg:person.013416235535.24
48 rdf:rest Nb6018b2126bc4178bad398f90ad0eac9
49 N9004fd75e70645aca244f3f13b13c2d6 rdf:first Ned176136fd6647f686f68b86444daebb
50 rdf:rest N0b27a5527f254c6e877df244a9fdd724
51 N95bb8b9006cb46d6b75b27c74b1a1d79 schema:name dimensions_id
52 schema:value pub.1084023566
53 rdf:type schema:PropertyValue
54 Nae8e2e9a95b64f3b9708f3c91757a400 schema:volumeNumber 22
55 rdf:type schema:PublicationVolume
56 Nb6018b2126bc4178bad398f90ad0eac9 rdf:first Nc187e44b7e4b4e14943bd2fa1691d521
57 rdf:rest N9004fd75e70645aca244f3f13b13c2d6
58 Nb637f1b5414b42aab9168c86c293f5f2 schema:name doi
59 schema:value 10.1007/s10015-017-0361-6
60 rdf:type schema:PropertyValue
61 Nc187e44b7e4b4e14943bd2fa1691d521 schema:affiliation https://www.grid.ac/institutes/grid.418251.b
62 schema:familyName Sakai
63 schema:givenName Yoshitaka
64 rdf:type schema:Person
65 Ned176136fd6647f686f68b86444daebb schema:affiliation https://www.grid.ac/institutes/grid.417547.4
66 schema:familyName Kadoya
67 schema:givenName Toru
68 rdf:type schema:Person
69 Nfe6d6ac0025944cbab5b57a273398e65 schema:name Springer Nature - SN SciGraph project
70 rdf:type schema:Organization
71 anzsrc-for:01 schema:inDefinedTermSet anzsrc-for:
72 schema:name Mathematical Sciences
73 rdf:type schema:DefinedTerm
74 anzsrc-for:0103 schema:inDefinedTermSet anzsrc-for:
75 schema:name Numerical and Computational Mathematics
76 rdf:type schema:DefinedTerm
77 sg:grant.5865312 http://pending.schema.org/fundedItem sg:pub.10.1007/s10015-017-0361-6
78 rdf:type schema:MonetaryGrant
79 sg:journal.1135365 schema:issn 1433-5298
80 1614-7456
81 schema:name Artificial Life and Robotics
82 rdf:type schema:Periodical
83 sg:person.010430333535.31 schema:affiliation https://www.grid.ac/institutes/grid.412339.e
84 schema:familyName Fukumoto
85 schema:givenName Hisao
86 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.010430333535.31
87 rdf:type schema:Person
88 sg:person.012705031237.88 schema:affiliation https://www.grid.ac/institutes/grid.412339.e
89 schema:familyName Wakuya
90 schema:givenName Hiroshi
91 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.012705031237.88
92 rdf:type schema:Person
93 sg:person.013416235535.24 schema:affiliation https://www.grid.ac/institutes/grid.412339.e
94 schema:familyName Itoh
95 schema:givenName Hideaki
96 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013416235535.24
97 rdf:type schema:Person
98 sg:person.014746150256.43 schema:affiliation https://www.grid.ac/institutes/grid.412339.e
99 schema:familyName Furukawa
100 schema:givenName Tatsuya
101 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014746150256.43
102 rdf:type schema:Person
103 sg:pub.10.1007/s10514-012-9315-y schema:sameAs https://app.dimensions.ai/details/publication/pub.1017190414
104 https://doi.org/10.1007/s10514-012-9315-y
105 rdf:type schema:CreativeWork
106 sg:pub.10.1007/s10994-012-5278-7 schema:sameAs https://app.dimensions.ai/details/publication/pub.1000956677
107 https://doi.org/10.1007/s10994-012-5278-7
108 rdf:type schema:CreativeWork
109 https://doi.org/10.1109/humanoids.2014.7041389 schema:sameAs https://app.dimensions.ai/details/publication/pub.1094167611
110 rdf:type schema:CreativeWork
111 https://doi.org/10.1145/1143844.1143963 schema:sameAs https://app.dimensions.ai/details/publication/pub.1029214029
112 rdf:type schema:CreativeWork
113 https://doi.org/10.1145/1553374.1553508 schema:sameAs https://app.dimensions.ai/details/publication/pub.1026898291
114 rdf:type schema:CreativeWork
115 https://doi.org/10.1177/0278364913482017 schema:sameAs https://app.dimensions.ai/details/publication/pub.1027688393
116 rdf:type schema:CreativeWork
117 https://doi.org/10.15607/rss.2012.viii.045 schema:sameAs https://app.dimensions.ai/details/publication/pub.1099108088
118 rdf:type schema:CreativeWork
119 https://doi.org/10.15607/rss.2012.viii.059 schema:sameAs https://app.dimensions.ai/details/publication/pub.1099108102
120 rdf:type schema:CreativeWork
121 https://doi.org/10.1613/jair.4649 schema:sameAs https://app.dimensions.ai/details/publication/pub.1105689999
122 rdf:type schema:CreativeWork
123 https://www.grid.ac/institutes/grid.412339.e schema:alternateName Saga University
124 schema:name Department of Electrical and Electronic Engineering, Graduate School of Science and Engineering, Saga University, 1 Honjo-machi, 840-8502, Saga, Japan
125 rdf:type schema:Organization
126 https://www.grid.ac/institutes/grid.417547.4 schema:alternateName Hitachi (Japan)
127 schema:name Hitachi Systems, Ltd., 1-2-1 Osaki, Sinagawa-ku, 141-8672, Tokyo, Japan
128 rdf:type schema:Organization
129 https://www.grid.ac/institutes/grid.418251.b schema:alternateName Fujitsu (Japan)
130 schema:name Fujitsu Limited, 1-5-2 Higashi-Shimbashi, Minato-ku, 105-7123, Tokyo, Japan
131 rdf:type schema:Organization
 




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


...