Fast incremental learning methods inspired by biological learning behavior View Full Text


Ontology type: schema:ScholarlyArticle     


Article Info

DATE

2005-07

AUTHORS

Koichiro Yamauchi, Takayuki Oohira, Takashi Omori

ABSTRACT

Model-based learning systems such as neural networks usually “forget” learned skills due to incremental learning of new instances. This is because the modification of a parameter interferes with old memories. Therefore, to avoid forgetting, incremental learning processes in these learning systems must include relearning of old instances. The relearning process, however, is time-consuming. We present two types of incremental learning method designed to achieve quick adaptation with low resources. One approach is to use a sleep phase to provide time for learning. The other one involves a “meta-learning module” that acquires learning skills through experience. The system carries out “reactive modification” of parameters not only to memorize new instances, but also to avoid forgetting old memories using a meta-learning module. More... »

PAGES

128-134

References to SciGraph publications

  • 2003-06-18. Meta-learning for Fast Incremental Learning in ARTIFICIAL NEURAL NETWORKS AND NEURAL INFORMATION PROCESSING — ICANN/ICONIP 2003
  • 1995-01. A new scheme for incremental learning in NEURAL PROCESSING LETTERS
  • Identifiers

    URI

    http://scigraph.springernature.com/pub.10.1007/s10015-004-0325-5

    DOI

    http://dx.doi.org/10.1007/s10015-004-0325-5

    DIMENSIONS

    https://app.dimensions.ai/details/publication/pub.1049622643


    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": "Graduate School of Information Science and Technology, Hokkaido University, Kita 14, Nishi 9, Kita-ku, 060-0814, Sapporo, Japan", 
              "id": "http://www.grid.ac/institutes/grid.39158.36", 
              "name": [
                "Graduate School of Information Science and Technology, Hokkaido University, Kita 14, Nishi 9, Kita-ku, 060-0814, Sapporo, Japan"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Yamauchi", 
            "givenName": "Koichiro", 
            "id": "sg:person.07364502315.92", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.07364502315.92"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Graduate School of Information Science and Technology, Hokkaido University, Kita 14, Nishi 9, Kita-ku, 060-0814, Sapporo, Japan", 
              "id": "http://www.grid.ac/institutes/grid.39158.36", 
              "name": [
                "Graduate School of Information Science and Technology, Hokkaido University, Kita 14, Nishi 9, Kita-ku, 060-0814, Sapporo, Japan"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Oohira", 
            "givenName": "Takayuki", 
            "id": "sg:person.011767412664.06", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.011767412664.06"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Graduate School of Information Science and Technology, Hokkaido University, Kita 14, Nishi 9, Kita-ku, 060-0814, Sapporo, Japan", 
              "id": "http://www.grid.ac/institutes/grid.39158.36", 
              "name": [
                "Graduate School of Information Science and Technology, Hokkaido University, Kita 14, Nishi 9, Kita-ku, 060-0814, Sapporo, Japan"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Omori", 
            "givenName": "Takashi", 
            "id": "sg:person.01263557346.07", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01263557346.07"
            ], 
            "type": "Person"
          }
        ], 
        "citation": [
          {
            "id": "sg:pub.10.1007/3-540-44989-2_20", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1010032192", 
              "https://doi.org/10.1007/3-540-44989-2_20"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/bf02312374", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1005936339", 
              "https://doi.org/10.1007/bf02312374"
            ], 
            "type": "CreativeWork"
          }
        ], 
        "datePublished": "2005-07", 
        "datePublishedReg": "2005-07-01", 
        "description": "Model-based learning systems such as neural networks usually \u201cforget\u201d learned skills due to incremental learning of new instances. This is because the modification of a parameter interferes with old memories. Therefore, to avoid forgetting, incremental learning processes in these learning systems must include relearning of old instances. The relearning process, however, is time-consuming. We present two types of incremental learning method designed to achieve quick adaptation with low resources. One approach is to use a sleep phase to provide time for learning. The other one involves a \u201cmeta-learning module\u201d that acquires learning skills through experience. The system carries out \u201creactive modification\u201d of parameters not only to memorize new instances, but also to avoid forgetting old memories using a meta-learning module.", 
        "genre": "article", 
        "id": "sg:pub.10.1007/s10015-004-0325-5", 
        "inLanguage": "en", 
        "isAccessibleForFree": false, 
        "isPartOf": [
          {
            "id": "sg:journal.1135365", 
            "issn": [
              "1433-5298", 
              "1614-7456"
            ], 
            "name": "Artificial Life and Robotics", 
            "publisher": "Springer Nature", 
            "type": "Periodical"
          }, 
          {
            "issueNumber": "3", 
            "type": "PublicationIssue"
          }, 
          {
            "type": "PublicationVolume", 
            "volumeNumber": "9"
          }
        ], 
        "keywords": [
          "old memories", 
          "model-based learning system", 
          "incremental learning method", 
          "incremental learning process", 
          "learning method", 
          "new instances", 
          "learning behavior", 
          "old instances", 
          "learning process", 
          "learning system", 
          "incremental learning", 
          "memory", 
          "neural network", 
          "sleep phase", 
          "skills", 
          "learning", 
          "quick adaptation", 
          "low resources", 
          "module", 
          "instances", 
          "system", 
          "experience", 
          "network", 
          "behavior", 
          "adaptation", 
          "resources", 
          "process", 
          "method", 
          "parameters", 
          "reactive modification", 
          "one", 
          "approach", 
          "time", 
          "types", 
          "modification", 
          "phase", 
          "meta-learning module", 
          "Fast incremental learning methods", 
          "biological learning behavior"
        ], 
        "name": "Fast incremental learning methods inspired by biological learning behavior", 
        "pagination": "128-134", 
        "productId": [
          {
            "name": "dimensions_id", 
            "type": "PropertyValue", 
            "value": [
              "pub.1049622643"
            ]
          }, 
          {
            "name": "doi", 
            "type": "PropertyValue", 
            "value": [
              "10.1007/s10015-004-0325-5"
            ]
          }
        ], 
        "sameAs": [
          "https://doi.org/10.1007/s10015-004-0325-5", 
          "https://app.dimensions.ai/details/publication/pub.1049622643"
        ], 
        "sdDataset": "articles", 
        "sdDatePublished": "2022-01-01T18:14", 
        "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_407.jsonl", 
        "type": "ScholarlyArticle", 
        "url": "https://doi.org/10.1007/s10015-004-0325-5"
      }
    ]
     

    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/s10015-004-0325-5'

    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/s10015-004-0325-5'

    Turtle is a human-readable linked data format.

    curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/s10015-004-0325-5'

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

    curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/s10015-004-0325-5'


     

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

    119 TRIPLES      22 PREDICATES      67 URIs      57 LITERALS      6 BLANK NODES

    Subject Predicate Object
    1 sg:pub.10.1007/s10015-004-0325-5 schema:about anzsrc-for:08
    2 anzsrc-for:0801
    3 schema:author N1bd3747e1f9948e88ffe1ebbedbb5c02
    4 schema:citation sg:pub.10.1007/3-540-44989-2_20
    5 sg:pub.10.1007/bf02312374
    6 schema:datePublished 2005-07
    7 schema:datePublishedReg 2005-07-01
    8 schema:description Model-based learning systems such as neural networks usually “forget” learned skills due to incremental learning of new instances. This is because the modification of a parameter interferes with old memories. Therefore, to avoid forgetting, incremental learning processes in these learning systems must include relearning of old instances. The relearning process, however, is time-consuming. We present two types of incremental learning method designed to achieve quick adaptation with low resources. One approach is to use a sleep phase to provide time for learning. The other one involves a “meta-learning module” that acquires learning skills through experience. The system carries out “reactive modification” of parameters not only to memorize new instances, but also to avoid forgetting old memories using a meta-learning module.
    9 schema:genre article
    10 schema:inLanguage en
    11 schema:isAccessibleForFree false
    12 schema:isPartOf N24e38aedd9e54f5694ae2420ae4131f2
    13 Nd8791362559c46f8ac659652f28502f5
    14 sg:journal.1135365
    15 schema:keywords Fast incremental learning methods
    16 adaptation
    17 approach
    18 behavior
    19 biological learning behavior
    20 experience
    21 incremental learning
    22 incremental learning method
    23 incremental learning process
    24 instances
    25 learning
    26 learning behavior
    27 learning method
    28 learning process
    29 learning system
    30 low resources
    31 memory
    32 meta-learning module
    33 method
    34 model-based learning system
    35 modification
    36 module
    37 network
    38 neural network
    39 new instances
    40 old instances
    41 old memories
    42 one
    43 parameters
    44 phase
    45 process
    46 quick adaptation
    47 reactive modification
    48 resources
    49 skills
    50 sleep phase
    51 system
    52 time
    53 types
    54 schema:name Fast incremental learning methods inspired by biological learning behavior
    55 schema:pagination 128-134
    56 schema:productId Nddb788ddabe44933abf8156f82de4328
    57 Nf734932cb8f24957b2dad6b1d2c8c84c
    58 schema:sameAs https://app.dimensions.ai/details/publication/pub.1049622643
    59 https://doi.org/10.1007/s10015-004-0325-5
    60 schema:sdDatePublished 2022-01-01T18:14
    61 schema:sdLicense https://scigraph.springernature.com/explorer/license/
    62 schema:sdPublisher N47673b66cb674ffca9fbf01c94d37e8b
    63 schema:url https://doi.org/10.1007/s10015-004-0325-5
    64 sgo:license sg:explorer/license/
    65 sgo:sdDataset articles
    66 rdf:type schema:ScholarlyArticle
    67 N1bd3747e1f9948e88ffe1ebbedbb5c02 rdf:first sg:person.07364502315.92
    68 rdf:rest Nb8859c8420a9456091a40677c60b893d
    69 N24e38aedd9e54f5694ae2420ae4131f2 schema:issueNumber 3
    70 rdf:type schema:PublicationIssue
    71 N47673b66cb674ffca9fbf01c94d37e8b schema:name Springer Nature - SN SciGraph project
    72 rdf:type schema:Organization
    73 N8a5a365662b24376946f4f7eda19a49b rdf:first sg:person.01263557346.07
    74 rdf:rest rdf:nil
    75 Nb8859c8420a9456091a40677c60b893d rdf:first sg:person.011767412664.06
    76 rdf:rest N8a5a365662b24376946f4f7eda19a49b
    77 Nd8791362559c46f8ac659652f28502f5 schema:volumeNumber 9
    78 rdf:type schema:PublicationVolume
    79 Nddb788ddabe44933abf8156f82de4328 schema:name doi
    80 schema:value 10.1007/s10015-004-0325-5
    81 rdf:type schema:PropertyValue
    82 Nf734932cb8f24957b2dad6b1d2c8c84c schema:name dimensions_id
    83 schema:value pub.1049622643
    84 rdf:type schema:PropertyValue
    85 anzsrc-for:08 schema:inDefinedTermSet anzsrc-for:
    86 schema:name Information and Computing Sciences
    87 rdf:type schema:DefinedTerm
    88 anzsrc-for:0801 schema:inDefinedTermSet anzsrc-for:
    89 schema:name Artificial Intelligence and Image Processing
    90 rdf:type schema:DefinedTerm
    91 sg:journal.1135365 schema:issn 1433-5298
    92 1614-7456
    93 schema:name Artificial Life and Robotics
    94 schema:publisher Springer Nature
    95 rdf:type schema:Periodical
    96 sg:person.011767412664.06 schema:affiliation grid-institutes:grid.39158.36
    97 schema:familyName Oohira
    98 schema:givenName Takayuki
    99 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.011767412664.06
    100 rdf:type schema:Person
    101 sg:person.01263557346.07 schema:affiliation grid-institutes:grid.39158.36
    102 schema:familyName Omori
    103 schema:givenName Takashi
    104 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01263557346.07
    105 rdf:type schema:Person
    106 sg:person.07364502315.92 schema:affiliation grid-institutes:grid.39158.36
    107 schema:familyName Yamauchi
    108 schema:givenName Koichiro
    109 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.07364502315.92
    110 rdf:type schema:Person
    111 sg:pub.10.1007/3-540-44989-2_20 schema:sameAs https://app.dimensions.ai/details/publication/pub.1010032192
    112 https://doi.org/10.1007/3-540-44989-2_20
    113 rdf:type schema:CreativeWork
    114 sg:pub.10.1007/bf02312374 schema:sameAs https://app.dimensions.ai/details/publication/pub.1005936339
    115 https://doi.org/10.1007/bf02312374
    116 rdf:type schema:CreativeWork
    117 grid-institutes:grid.39158.36 schema:alternateName Graduate School of Information Science and Technology, Hokkaido University, Kita 14, Nishi 9, Kita-ku, 060-0814, Sapporo, Japan
    118 schema:name Graduate School of Information Science and Technology, Hokkaido University, Kita 14, Nishi 9, Kita-ku, 060-0814, Sapporo, Japan
    119 rdf:type schema:Organization
     




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


    ...