INFORMATION PROCESSING APPARATUS, METHOD FOR CONTROLLING INFORMATION PROCESSING APPARATUS, AND PROGRAM

To construct a learned model composed of a multilayer neural network in a more suitable mode.SOLUTION: An initial feature conversion unit 101 generates, from input data, a feature quantity indicating the feature of the input data. A feature conversion unit 102 sequentially and gradually executes a p...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
1. Verfasser: YAZAWA TOMONORI
Format: Patent
Sprache:eng ; jpn
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator YAZAWA TOMONORI
description To construct a learned model composed of a multilayer neural network in a more suitable mode.SOLUTION: An initial feature conversion unit 101 generates, from input data, a feature quantity indicating the feature of the input data. A feature conversion unit 102 sequentially and gradually executes a plurality of types of arithmetic processing on the generated feature quantity to convert the generated feature quantity into another feature quantity. The first arithmetic processing being at least part of the plurality of types of arithmetic processing takes, as input, a plurality of first feature quantities output as results of a plurality of types of second arithmetic processing earlier in the processing order than the first arithmetic processing by the minimum processing order interval or more which is set in advance, and applies bijection of the feature quantities to at least a partial feature quantity of the plurality of first feature quantities, and thereby generates a second feature quantity. The minimum processing order interval is 2 or more.SELECTED DRAWING: Figure 2 【課題】多層階のニューラルネットワークにより構成された学習済モデルの構築をより好適な態様で実現する。【解決手段】初期特徴変換部101は、入力データから、当該入力データの特徴を示した特徴量を生成する。特徴変換部102は、生成された前記特徴量に対して、複数の演算処理を段階的に順次実行することで他の特徴量に変換する。上記複数の演算処理のうち少なくとも一部の第1の演算処理は、当該第1の演算処理よりも処理順序があらかじめ設定された最小処理順序間隔以上前の複数の第2の演算処理それぞれの結果として出力される複数の第1の特徴量を入力として、当該複数の第1の特徴量のうちの少なくとも一部の特徴量を対象として、当該特徴量の全単射を適用して、第2の特徴量を生成する。上記最小処理順序間隔は2以上である。【選択図】図2
format Patent
fullrecord <record><control><sourceid>epo_EVB</sourceid><recordid>TN_cdi_epo_espacenet_JP2024017739A</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>JP2024017739A</sourcerecordid><originalsourceid>FETCH-epo_espacenet_JP2024017739A3</originalsourceid><addsrcrecordid>eNrjZEjz9HPzD_J1DPH091MICPJ3dg0O9vRzV3AMCHAMcgwJDdZR8HUN8fB3UQAqU3D29wsJ8vfxAakgrNHRzwUk4x7k6MvDwJqWmFOcyguluRmU3FxDnD10Uwvy41OLCxKTU_NSS-K9AowMjEwMDM3NjS0djYlSBACBdDYl</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>patent</recordtype></control><display><type>patent</type><title>INFORMATION PROCESSING APPARATUS, METHOD FOR CONTROLLING INFORMATION PROCESSING APPARATUS, AND PROGRAM</title><source>esp@cenet</source><creator>YAZAWA TOMONORI</creator><creatorcontrib>YAZAWA TOMONORI</creatorcontrib><description>To construct a learned model composed of a multilayer neural network in a more suitable mode.SOLUTION: An initial feature conversion unit 101 generates, from input data, a feature quantity indicating the feature of the input data. A feature conversion unit 102 sequentially and gradually executes a plurality of types of arithmetic processing on the generated feature quantity to convert the generated feature quantity into another feature quantity. The first arithmetic processing being at least part of the plurality of types of arithmetic processing takes, as input, a plurality of first feature quantities output as results of a plurality of types of second arithmetic processing earlier in the processing order than the first arithmetic processing by the minimum processing order interval or more which is set in advance, and applies bijection of the feature quantities to at least a partial feature quantity of the plurality of first feature quantities, and thereby generates a second feature quantity. The minimum processing order interval is 2 or more.SELECTED DRAWING: Figure 2 【課題】多層階のニューラルネットワークにより構成された学習済モデルの構築をより好適な態様で実現する。【解決手段】初期特徴変換部101は、入力データから、当該入力データの特徴を示した特徴量を生成する。特徴変換部102は、生成された前記特徴量に対して、複数の演算処理を段階的に順次実行することで他の特徴量に変換する。上記複数の演算処理のうち少なくとも一部の第1の演算処理は、当該第1の演算処理よりも処理順序があらかじめ設定された最小処理順序間隔以上前の複数の第2の演算処理それぞれの結果として出力される複数の第1の特徴量を入力として、当該複数の第1の特徴量のうちの少なくとも一部の特徴量を対象として、当該特徴量の全単射を適用して、第2の特徴量を生成する。上記最小処理順序間隔は2以上である。【選択図】図2</description><language>eng ; jpn</language><subject>CALCULATING ; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS ; COMPUTING ; COUNTING ; PHYSICS</subject><creationdate>2024</creationdate><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&amp;date=20240208&amp;DB=EPODOC&amp;CC=JP&amp;NR=2024017739A$$EHTML$$P50$$Gepo$$Hfree_for_read</linktohtml><link.rule.ids>230,308,776,881,25542,76516</link.rule.ids><linktorsrc>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&amp;date=20240208&amp;DB=EPODOC&amp;CC=JP&amp;NR=2024017739A$$EView_record_in_European_Patent_Office$$FView_record_in_$$GEuropean_Patent_Office$$Hfree_for_read</linktorsrc></links><search><creatorcontrib>YAZAWA TOMONORI</creatorcontrib><title>INFORMATION PROCESSING APPARATUS, METHOD FOR CONTROLLING INFORMATION PROCESSING APPARATUS, AND PROGRAM</title><description>To construct a learned model composed of a multilayer neural network in a more suitable mode.SOLUTION: An initial feature conversion unit 101 generates, from input data, a feature quantity indicating the feature of the input data. A feature conversion unit 102 sequentially and gradually executes a plurality of types of arithmetic processing on the generated feature quantity to convert the generated feature quantity into another feature quantity. The first arithmetic processing being at least part of the plurality of types of arithmetic processing takes, as input, a plurality of first feature quantities output as results of a plurality of types of second arithmetic processing earlier in the processing order than the first arithmetic processing by the minimum processing order interval or more which is set in advance, and applies bijection of the feature quantities to at least a partial feature quantity of the plurality of first feature quantities, and thereby generates a second feature quantity. The minimum processing order interval is 2 or more.SELECTED DRAWING: Figure 2 【課題】多層階のニューラルネットワークにより構成された学習済モデルの構築をより好適な態様で実現する。【解決手段】初期特徴変換部101は、入力データから、当該入力データの特徴を示した特徴量を生成する。特徴変換部102は、生成された前記特徴量に対して、複数の演算処理を段階的に順次実行することで他の特徴量に変換する。上記複数の演算処理のうち少なくとも一部の第1の演算処理は、当該第1の演算処理よりも処理順序があらかじめ設定された最小処理順序間隔以上前の複数の第2の演算処理それぞれの結果として出力される複数の第1の特徴量を入力として、当該複数の第1の特徴量のうちの少なくとも一部の特徴量を対象として、当該特徴量の全単射を適用して、第2の特徴量を生成する。上記最小処理順序間隔は2以上である。【選択図】図2</description><subject>CALCULATING</subject><subject>COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS</subject><subject>COMPUTING</subject><subject>COUNTING</subject><subject>PHYSICS</subject><fulltext>true</fulltext><rsrctype>patent</rsrctype><creationdate>2024</creationdate><recordtype>patent</recordtype><sourceid>EVB</sourceid><recordid>eNrjZEjz9HPzD_J1DPH091MICPJ3dg0O9vRzV3AMCHAMcgwJDdZR8HUN8fB3UQAqU3D29wsJ8vfxAakgrNHRzwUk4x7k6MvDwJqWmFOcyguluRmU3FxDnD10Uwvy41OLCxKTU_NSS-K9AowMjEwMDM3NjS0djYlSBACBdDYl</recordid><startdate>20240208</startdate><enddate>20240208</enddate><creator>YAZAWA TOMONORI</creator><scope>EVB</scope></search><sort><creationdate>20240208</creationdate><title>INFORMATION PROCESSING APPARATUS, METHOD FOR CONTROLLING INFORMATION PROCESSING APPARATUS, AND PROGRAM</title><author>YAZAWA TOMONORI</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-epo_espacenet_JP2024017739A3</frbrgroupid><rsrctype>patents</rsrctype><prefilter>patents</prefilter><language>eng ; jpn</language><creationdate>2024</creationdate><topic>CALCULATING</topic><topic>COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS</topic><topic>COMPUTING</topic><topic>COUNTING</topic><topic>PHYSICS</topic><toplevel>online_resources</toplevel><creatorcontrib>YAZAWA TOMONORI</creatorcontrib><collection>esp@cenet</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>YAZAWA TOMONORI</au><format>patent</format><genre>patent</genre><ristype>GEN</ristype><title>INFORMATION PROCESSING APPARATUS, METHOD FOR CONTROLLING INFORMATION PROCESSING APPARATUS, AND PROGRAM</title><date>2024-02-08</date><risdate>2024</risdate><abstract>To construct a learned model composed of a multilayer neural network in a more suitable mode.SOLUTION: An initial feature conversion unit 101 generates, from input data, a feature quantity indicating the feature of the input data. A feature conversion unit 102 sequentially and gradually executes a plurality of types of arithmetic processing on the generated feature quantity to convert the generated feature quantity into another feature quantity. The first arithmetic processing being at least part of the plurality of types of arithmetic processing takes, as input, a plurality of first feature quantities output as results of a plurality of types of second arithmetic processing earlier in the processing order than the first arithmetic processing by the minimum processing order interval or more which is set in advance, and applies bijection of the feature quantities to at least a partial feature quantity of the plurality of first feature quantities, and thereby generates a second feature quantity. The minimum processing order interval is 2 or more.SELECTED DRAWING: Figure 2 【課題】多層階のニューラルネットワークにより構成された学習済モデルの構築をより好適な態様で実現する。【解決手段】初期特徴変換部101は、入力データから、当該入力データの特徴を示した特徴量を生成する。特徴変換部102は、生成された前記特徴量に対して、複数の演算処理を段階的に順次実行することで他の特徴量に変換する。上記複数の演算処理のうち少なくとも一部の第1の演算処理は、当該第1の演算処理よりも処理順序があらかじめ設定された最小処理順序間隔以上前の複数の第2の演算処理それぞれの結果として出力される複数の第1の特徴量を入力として、当該複数の第1の特徴量のうちの少なくとも一部の特徴量を対象として、当該特徴量の全単射を適用して、第2の特徴量を生成する。上記最小処理順序間隔は2以上である。【選択図】図2</abstract><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier
ispartof
issn
language eng ; jpn
recordid cdi_epo_espacenet_JP2024017739A
source esp@cenet
subjects CALCULATING
COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
COMPUTING
COUNTING
PHYSICS
title INFORMATION PROCESSING APPARATUS, METHOD FOR CONTROLLING INFORMATION PROCESSING APPARATUS, AND PROGRAM
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-13T22%3A00%3A04IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-epo_EVB&rft_val_fmt=info:ofi/fmt:kev:mtx:patent&rft.genre=patent&rft.au=YAZAWA%20TOMONORI&rft.date=2024-02-08&rft_id=info:doi/&rft_dat=%3Cepo_EVB%3EJP2024017739A%3C/epo_EVB%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true