Variants of residual smoothing with a small residual gap View Full Text


Ontology type: schema:ScholarlyArticle     


Article Info

DATE

2019-04-02

AUTHORS

Kensuke Aihara, Ryosuke Komeyama, Emiko Ishiwata

ABSTRACT

Short-recurrence Krylov subspace methods, such as conjugate gradient squared-type methods, often exhibit large oscillations in the residual norms, leading to a large residual gap and a loss of attainable accuracy for the approximate solutions. Residual smoothing is useful for obtaining smooth convergence for the residual norms, but it has been shown that this does not improve the maximum attainable accuracy in most cases. In the present study, we reformulate the smoothing scheme from a novel perspective. The smoothed sequences do not usually affect the primary sequences in conventional smoothing schemes. In contrast, we design a variant of residual smoothing in which the primary and smoothed sequences influence each other. This approach enables us to avoid the propagation of large rounding errors, and results in a smaller residual gap, and thus a higher attainable accuracy. We present a rounding error analysis and numerical experiments to demonstrate the effectiveness of our proposed smoothing scheme. More... »

PAGES

1-20

Journal

TITLE

BIT Numerical Mathematics

ISSUE

N/A

VOLUME

N/A

From Grant

  • Identifiers

    URI

    http://scigraph.springernature.com/pub.10.1007/s10543-019-00751-w

    DOI

    http://dx.doi.org/10.1007/s10543-019-00751-w

    DIMENSIONS

    https://app.dimensions.ai/details/publication/pub.1113199724


    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/2008/0104", 
            "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
            "name": "Statistics", 
            "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": "Tokyo City University", 
              "id": "https://www.grid.ac/institutes/grid.458395.6", 
              "name": [
                "Department of Computer Science, Tokyo City University, 1-28-1 Tamazutsumi, Setagaya-ku, 158-8557, Tokyo, Japan"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Aihara", 
            "givenName": "Kensuke", 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "name": [
                "Hitachi Industry & Control Solutions, Ltd., 5-1-26 Omika-cho, 319-1221, Hitachi-shi, Ibaraki, Japan"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Komeyama", 
            "givenName": "Ryosuke", 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Tokyo University of Science", 
              "id": "https://www.grid.ac/institutes/grid.143643.7", 
              "name": [
                "Department of Applied Mathematics, Tokyo University of Science, 1-3 Kagurazaka, Shinjuku-ku, 162-8601, Tokyo, Japan"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Ishiwata", 
            "givenName": "Emiko", 
            "type": "Person"
          }
        ], 
        "citation": [
          {
            "id": "sg:pub.10.1007/s11075-016-0183-y", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1000665224", 
              "https://doi.org/10.1007/s11075-016-0183-y"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/s11075-016-0183-y", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1000665224", 
              "https://doi.org/10.1007/s11075-016-0183-y"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/bf02309342", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1001644990", 
              "https://doi.org/10.1007/bf02309342"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/bf02309342", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1001644990", 
              "https://doi.org/10.1007/bf02309342"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1023/a:1011889705659", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1006896721", 
              "https://doi.org/10.1023/a:1011889705659"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1023/a:1021917801600", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1007607749", 
              "https://doi.org/10.1023/a:1021917801600"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1016/0377-0427(95)00227-8", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1015853065"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1145/2049662.2049663", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1021694365"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/bf01385726", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1049684739", 
              "https://doi.org/10.1007/bf01385726"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/bfb0080116", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1050935180", 
              "https://doi.org/10.1007/bfb0080116"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1137/0712047", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1062852294"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1137/0910004", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1062857216"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1137/0913035", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1062857471"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1137/0914029", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1062857550"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1137/0915021", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1062857628"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1137/0915023", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1062857630"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1137/s0895479802403459", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1062881765"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1137/s0895479895284944", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1062882166"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1137/s1064827599353865", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1062884714"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1137/1.9780898718003", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1098555810"
            ], 
            "type": "CreativeWork"
          }
        ], 
        "datePublished": "2019-04-02", 
        "datePublishedReg": "2019-04-02", 
        "description": "Short-recurrence Krylov subspace methods, such as conjugate gradient squared-type methods, often exhibit large oscillations in the residual norms, leading to a large residual gap and a loss of attainable accuracy for the approximate solutions. Residual smoothing is useful for obtaining smooth convergence for the residual norms, but it has been shown that this does not improve the maximum attainable accuracy in most cases. In the present study, we reformulate the smoothing scheme from a novel perspective. The smoothed sequences do not usually affect the primary sequences in conventional smoothing schemes. In contrast, we design a variant of residual smoothing in which the primary and smoothed sequences influence each other. This approach enables us to avoid the propagation of large rounding errors, and results in a smaller residual gap, and thus a higher attainable accuracy. We present a rounding error analysis and numerical experiments to demonstrate the effectiveness of our proposed smoothing scheme.", 
        "genre": "research_article", 
        "id": "sg:pub.10.1007/s10543-019-00751-w", 
        "inLanguage": [
          "en"
        ], 
        "isAccessibleForFree": false, 
        "isFundedItemOf": [
          {
            "id": "sg:grant.7544894", 
            "type": "MonetaryGrant"
          }
        ], 
        "isPartOf": [
          {
            "id": "sg:journal.1136252", 
            "issn": [
              "0006-3835", 
              "1572-9125"
            ], 
            "name": "BIT Numerical Mathematics", 
            "type": "Periodical"
          }
        ], 
        "name": "Variants of residual smoothing with a small residual gap", 
        "pagination": "1-20", 
        "productId": [
          {
            "name": "doi", 
            "type": "PropertyValue", 
            "value": [
              "10.1007/s10543-019-00751-w"
            ]
          }, 
          {
            "name": "readcube_id", 
            "type": "PropertyValue", 
            "value": [
              "6db3e37a8ae32bfb558aa71f28da7e5a3d3427c7c633442ef1cd23748e5fd740"
            ]
          }, 
          {
            "name": "dimensions_id", 
            "type": "PropertyValue", 
            "value": [
              "pub.1113199724"
            ]
          }
        ], 
        "sameAs": [
          "https://doi.org/10.1007/s10543-019-00751-w", 
          "https://app.dimensions.ai/details/publication/pub.1113199724"
        ], 
        "sdDataset": "articles", 
        "sdDatePublished": "2019-04-15T09:23", 
        "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/v90/0000000376_0000000376/records_56185_00000006.jsonl", 
        "type": "ScholarlyArticle", 
        "url": "https://link.springer.com/10.1007%2Fs10543-019-00751-w"
      }
    ]
     

    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/s10543-019-00751-w'

    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/s10543-019-00751-w'

    Turtle is a human-readable linked data format.

    curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/s10543-019-00751-w'

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

    curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/s10543-019-00751-w'


     

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

    133 TRIPLES      21 PREDICATES      42 URIs      16 LITERALS      5 BLANK NODES

    Subject Predicate Object
    1 sg:pub.10.1007/s10543-019-00751-w schema:about anzsrc-for:01
    2 anzsrc-for:0104
    3 schema:author N48e68c408b734a59ba8c59696140e92b
    4 schema:citation sg:pub.10.1007/bf01385726
    5 sg:pub.10.1007/bf02309342
    6 sg:pub.10.1007/bfb0080116
    7 sg:pub.10.1007/s11075-016-0183-y
    8 sg:pub.10.1023/a:1011889705659
    9 sg:pub.10.1023/a:1021917801600
    10 https://doi.org/10.1016/0377-0427(95)00227-8
    11 https://doi.org/10.1137/0712047
    12 https://doi.org/10.1137/0910004
    13 https://doi.org/10.1137/0913035
    14 https://doi.org/10.1137/0914029
    15 https://doi.org/10.1137/0915021
    16 https://doi.org/10.1137/0915023
    17 https://doi.org/10.1137/1.9780898718003
    18 https://doi.org/10.1137/s0895479802403459
    19 https://doi.org/10.1137/s0895479895284944
    20 https://doi.org/10.1137/s1064827599353865
    21 https://doi.org/10.1145/2049662.2049663
    22 schema:datePublished 2019-04-02
    23 schema:datePublishedReg 2019-04-02
    24 schema:description Short-recurrence Krylov subspace methods, such as conjugate gradient squared-type methods, often exhibit large oscillations in the residual norms, leading to a large residual gap and a loss of attainable accuracy for the approximate solutions. Residual smoothing is useful for obtaining smooth convergence for the residual norms, but it has been shown that this does not improve the maximum attainable accuracy in most cases. In the present study, we reformulate the smoothing scheme from a novel perspective. The smoothed sequences do not usually affect the primary sequences in conventional smoothing schemes. In contrast, we design a variant of residual smoothing in which the primary and smoothed sequences influence each other. This approach enables us to avoid the propagation of large rounding errors, and results in a smaller residual gap, and thus a higher attainable accuracy. We present a rounding error analysis and numerical experiments to demonstrate the effectiveness of our proposed smoothing scheme.
    25 schema:genre research_article
    26 schema:inLanguage en
    27 schema:isAccessibleForFree false
    28 schema:isPartOf sg:journal.1136252
    29 schema:name Variants of residual smoothing with a small residual gap
    30 schema:pagination 1-20
    31 schema:productId N3d2aec2b10a24e2f8eaff9ac7c4b28df
    32 N400fce5fd8cb438096181dc94d51ba97
    33 Nedfaa4abec1a45afa963cbaf1eaf6c7f
    34 schema:sameAs https://app.dimensions.ai/details/publication/pub.1113199724
    35 https://doi.org/10.1007/s10543-019-00751-w
    36 schema:sdDatePublished 2019-04-15T09:23
    37 schema:sdLicense https://scigraph.springernature.com/explorer/license/
    38 schema:sdPublisher Nea5aafcbc5af4144a51660b723a28652
    39 schema:url https://link.springer.com/10.1007%2Fs10543-019-00751-w
    40 sgo:license sg:explorer/license/
    41 sgo:sdDataset articles
    42 rdf:type schema:ScholarlyArticle
    43 N2a62115f136e420db69896da7b9c956f schema:name Hitachi Industry & Control Solutions, Ltd., 5-1-26 Omika-cho, 319-1221, Hitachi-shi, Ibaraki, Japan
    44 rdf:type schema:Organization
    45 N3d2aec2b10a24e2f8eaff9ac7c4b28df schema:name dimensions_id
    46 schema:value pub.1113199724
    47 rdf:type schema:PropertyValue
    48 N400fce5fd8cb438096181dc94d51ba97 schema:name doi
    49 schema:value 10.1007/s10543-019-00751-w
    50 rdf:type schema:PropertyValue
    51 N460123a82e154b8db35007754230dc5e rdf:first Nf39b575fcd854547afb4cf2782085eed
    52 rdf:rest rdf:nil
    53 N48e68c408b734a59ba8c59696140e92b rdf:first N96a60798d0fb41209302e0083c0ccc2c
    54 rdf:rest Nef6e8c78468d490d9d8621358af622f7
    55 N96a60798d0fb41209302e0083c0ccc2c schema:affiliation https://www.grid.ac/institutes/grid.458395.6
    56 schema:familyName Aihara
    57 schema:givenName Kensuke
    58 rdf:type schema:Person
    59 Na3261a6da4cd4e2c9391fdf5a0212df6 schema:affiliation N2a62115f136e420db69896da7b9c956f
    60 schema:familyName Komeyama
    61 schema:givenName Ryosuke
    62 rdf:type schema:Person
    63 Nea5aafcbc5af4144a51660b723a28652 schema:name Springer Nature - SN SciGraph project
    64 rdf:type schema:Organization
    65 Nedfaa4abec1a45afa963cbaf1eaf6c7f schema:name readcube_id
    66 schema:value 6db3e37a8ae32bfb558aa71f28da7e5a3d3427c7c633442ef1cd23748e5fd740
    67 rdf:type schema:PropertyValue
    68 Nef6e8c78468d490d9d8621358af622f7 rdf:first Na3261a6da4cd4e2c9391fdf5a0212df6
    69 rdf:rest N460123a82e154b8db35007754230dc5e
    70 Nf39b575fcd854547afb4cf2782085eed schema:affiliation https://www.grid.ac/institutes/grid.143643.7
    71 schema:familyName Ishiwata
    72 schema:givenName Emiko
    73 rdf:type schema:Person
    74 anzsrc-for:01 schema:inDefinedTermSet anzsrc-for:
    75 schema:name Mathematical Sciences
    76 rdf:type schema:DefinedTerm
    77 anzsrc-for:0104 schema:inDefinedTermSet anzsrc-for:
    78 schema:name Statistics
    79 rdf:type schema:DefinedTerm
    80 sg:grant.7544894 http://pending.schema.org/fundedItem sg:pub.10.1007/s10543-019-00751-w
    81 rdf:type schema:MonetaryGrant
    82 sg:journal.1136252 schema:issn 0006-3835
    83 1572-9125
    84 schema:name BIT Numerical Mathematics
    85 rdf:type schema:Periodical
    86 sg:pub.10.1007/bf01385726 schema:sameAs https://app.dimensions.ai/details/publication/pub.1049684739
    87 https://doi.org/10.1007/bf01385726
    88 rdf:type schema:CreativeWork
    89 sg:pub.10.1007/bf02309342 schema:sameAs https://app.dimensions.ai/details/publication/pub.1001644990
    90 https://doi.org/10.1007/bf02309342
    91 rdf:type schema:CreativeWork
    92 sg:pub.10.1007/bfb0080116 schema:sameAs https://app.dimensions.ai/details/publication/pub.1050935180
    93 https://doi.org/10.1007/bfb0080116
    94 rdf:type schema:CreativeWork
    95 sg:pub.10.1007/s11075-016-0183-y schema:sameAs https://app.dimensions.ai/details/publication/pub.1000665224
    96 https://doi.org/10.1007/s11075-016-0183-y
    97 rdf:type schema:CreativeWork
    98 sg:pub.10.1023/a:1011889705659 schema:sameAs https://app.dimensions.ai/details/publication/pub.1006896721
    99 https://doi.org/10.1023/a:1011889705659
    100 rdf:type schema:CreativeWork
    101 sg:pub.10.1023/a:1021917801600 schema:sameAs https://app.dimensions.ai/details/publication/pub.1007607749
    102 https://doi.org/10.1023/a:1021917801600
    103 rdf:type schema:CreativeWork
    104 https://doi.org/10.1016/0377-0427(95)00227-8 schema:sameAs https://app.dimensions.ai/details/publication/pub.1015853065
    105 rdf:type schema:CreativeWork
    106 https://doi.org/10.1137/0712047 schema:sameAs https://app.dimensions.ai/details/publication/pub.1062852294
    107 rdf:type schema:CreativeWork
    108 https://doi.org/10.1137/0910004 schema:sameAs https://app.dimensions.ai/details/publication/pub.1062857216
    109 rdf:type schema:CreativeWork
    110 https://doi.org/10.1137/0913035 schema:sameAs https://app.dimensions.ai/details/publication/pub.1062857471
    111 rdf:type schema:CreativeWork
    112 https://doi.org/10.1137/0914029 schema:sameAs https://app.dimensions.ai/details/publication/pub.1062857550
    113 rdf:type schema:CreativeWork
    114 https://doi.org/10.1137/0915021 schema:sameAs https://app.dimensions.ai/details/publication/pub.1062857628
    115 rdf:type schema:CreativeWork
    116 https://doi.org/10.1137/0915023 schema:sameAs https://app.dimensions.ai/details/publication/pub.1062857630
    117 rdf:type schema:CreativeWork
    118 https://doi.org/10.1137/1.9780898718003 schema:sameAs https://app.dimensions.ai/details/publication/pub.1098555810
    119 rdf:type schema:CreativeWork
    120 https://doi.org/10.1137/s0895479802403459 schema:sameAs https://app.dimensions.ai/details/publication/pub.1062881765
    121 rdf:type schema:CreativeWork
    122 https://doi.org/10.1137/s0895479895284944 schema:sameAs https://app.dimensions.ai/details/publication/pub.1062882166
    123 rdf:type schema:CreativeWork
    124 https://doi.org/10.1137/s1064827599353865 schema:sameAs https://app.dimensions.ai/details/publication/pub.1062884714
    125 rdf:type schema:CreativeWork
    126 https://doi.org/10.1145/2049662.2049663 schema:sameAs https://app.dimensions.ai/details/publication/pub.1021694365
    127 rdf:type schema:CreativeWork
    128 https://www.grid.ac/institutes/grid.143643.7 schema:alternateName Tokyo University of Science
    129 schema:name Department of Applied Mathematics, Tokyo University of Science, 1-3 Kagurazaka, Shinjuku-ku, 162-8601, Tokyo, Japan
    130 rdf:type schema:Organization
    131 https://www.grid.ac/institutes/grid.458395.6 schema:alternateName Tokyo City University
    132 schema:name Department of Computer Science, Tokyo City University, 1-28-1 Tamazutsumi, Setagaya-ku, 158-8557, Tokyo, Japan
    133 rdf:type schema:Organization
     




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


    ...