INFORMATION PROCESSING APPARATUS AND INFORMATION PROCESSING METHOD

To provide an information processing apparatus configured to effectively increase the speed of learning using DNN regardless of learning methods, and an information processing method.SOLUTION: An information processing apparatus includes a learning unit which executes learning using a neural network...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
1. Verfasser: MIKAMI HIROAKI
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 MIKAMI HIROAKI
description To provide an information processing apparatus configured to effectively increase the speed of learning using DNN regardless of learning methods, and an information processing method.SOLUTION: An information processing apparatus includes a learning unit which executes learning using a neural network. The learning unit dynamically modifies a value of batch size during learning on the basis of a gap value with an ideal state relating to learning output by the neural network. A batch size modifying unit dynamically modifies a value of batch size during the learning, which uses a neural network, on the basis of a gap value with an ideal state relating to learning output by the neural network.SELECTED DRAWING: Figure 3 【課題】DNNによる学習を学習手法に依らず効果的に高速化する情報処理装置および情報処理方法を提供する。【解決手段】情報処理装置は、ニューラルネットワークを用いた学習を行う学習部を備える。学習部は、ニューラルネットワークが出力する学習に係る理想状態とのギャップ値に基づいて、学習中にバッチサイズの値を動的に変更する。また、ニューラルネットワークを用いた学習を行うことを含み、バッチサイズ変更部は、ニューラルネットワークが出力する学習に係る理想状態とのギャップ値に基づいて、学習中にバッチサイズの値を動的に変更する。【選択図】図3
format Patent
fullrecord <record><control><sourceid>epo_EVB</sourceid><recordid>TN_cdi_epo_espacenet_JP2020042591A</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>JP2020042591A</sourcerecordid><originalsourceid>FETCH-epo_espacenet_JP2020042591A3</originalsourceid><addsrcrecordid>eNrjZHDy9HPzD_J1DPH091MICPJ3dg0O9vRzV3AMCHAMcgwJDVZw9HNRwKHI1zXEw9-Fh4E1LTGnOJUXSnMzKLm5hjh76KYW5MenFhckJqfmpZbEewUYGRgZGJgYmVoaOhoTpQgAluUsGg</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>patent</recordtype></control><display><type>patent</type><title>INFORMATION PROCESSING APPARATUS AND INFORMATION PROCESSING METHOD</title><source>esp@cenet</source><creator>MIKAMI HIROAKI</creator><creatorcontrib>MIKAMI HIROAKI</creatorcontrib><description>To provide an information processing apparatus configured to effectively increase the speed of learning using DNN regardless of learning methods, and an information processing method.SOLUTION: An information processing apparatus includes a learning unit which executes learning using a neural network. The learning unit dynamically modifies a value of batch size during learning on the basis of a gap value with an ideal state relating to learning output by the neural network. A batch size modifying unit dynamically modifies a value of batch size during the learning, which uses a neural network, on the basis of a gap value with an ideal state relating to learning output by the neural network.SELECTED DRAWING: Figure 3 【課題】DNNによる学習を学習手法に依らず効果的に高速化する情報処理装置および情報処理方法を提供する。【解決手段】情報処理装置は、ニューラルネットワークを用いた学習を行う学習部を備える。学習部は、ニューラルネットワークが出力する学習に係る理想状態とのギャップ値に基づいて、学習中にバッチサイズの値を動的に変更する。また、ニューラルネットワークを用いた学習を行うことを含み、バッチサイズ変更部は、ニューラルネットワークが出力する学習に係る理想状態とのギャップ値に基づいて、学習中にバッチサイズの値を動的に変更する。【選択図】図3</description><language>eng ; jpn</language><subject>CALCULATING ; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS ; COMPUTING ; COUNTING ; PHYSICS</subject><creationdate>2020</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=20200319&amp;DB=EPODOC&amp;CC=JP&amp;NR=2020042591A$$EHTML$$P50$$Gepo$$Hfree_for_read</linktohtml><link.rule.ids>230,309,782,887,25571,76555</link.rule.ids><linktorsrc>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&amp;date=20200319&amp;DB=EPODOC&amp;CC=JP&amp;NR=2020042591A$$EView_record_in_European_Patent_Office$$FView_record_in_$$GEuropean_Patent_Office$$Hfree_for_read</linktorsrc></links><search><creatorcontrib>MIKAMI HIROAKI</creatorcontrib><title>INFORMATION PROCESSING APPARATUS AND INFORMATION PROCESSING METHOD</title><description>To provide an information processing apparatus configured to effectively increase the speed of learning using DNN regardless of learning methods, and an information processing method.SOLUTION: An information processing apparatus includes a learning unit which executes learning using a neural network. The learning unit dynamically modifies a value of batch size during learning on the basis of a gap value with an ideal state relating to learning output by the neural network. A batch size modifying unit dynamically modifies a value of batch size during the learning, which uses a neural network, on the basis of a gap value with an ideal state relating to learning output by the neural network.SELECTED DRAWING: Figure 3 【課題】DNNによる学習を学習手法に依らず効果的に高速化する情報処理装置および情報処理方法を提供する。【解決手段】情報処理装置は、ニューラルネットワークを用いた学習を行う学習部を備える。学習部は、ニューラルネットワークが出力する学習に係る理想状態とのギャップ値に基づいて、学習中にバッチサイズの値を動的に変更する。また、ニューラルネットワークを用いた学習を行うことを含み、バッチサイズ変更部は、ニューラルネットワークが出力する学習に係る理想状態とのギャップ値に基づいて、学習中にバッチサイズの値を動的に変更する。【選択図】図3</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>2020</creationdate><recordtype>patent</recordtype><sourceid>EVB</sourceid><recordid>eNrjZHDy9HPzD_J1DPH091MICPJ3dg0O9vRzV3AMCHAMcgwJDVZw9HNRwKHI1zXEw9-Fh4E1LTGnOJUXSnMzKLm5hjh76KYW5MenFhckJqfmpZbEewUYGRgZGJgYmVoaOhoTpQgAluUsGg</recordid><startdate>20200319</startdate><enddate>20200319</enddate><creator>MIKAMI HIROAKI</creator><scope>EVB</scope></search><sort><creationdate>20200319</creationdate><title>INFORMATION PROCESSING APPARATUS AND INFORMATION PROCESSING METHOD</title><author>MIKAMI HIROAKI</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-epo_espacenet_JP2020042591A3</frbrgroupid><rsrctype>patents</rsrctype><prefilter>patents</prefilter><language>eng ; jpn</language><creationdate>2020</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>MIKAMI HIROAKI</creatorcontrib><collection>esp@cenet</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>MIKAMI HIROAKI</au><format>patent</format><genre>patent</genre><ristype>GEN</ristype><title>INFORMATION PROCESSING APPARATUS AND INFORMATION PROCESSING METHOD</title><date>2020-03-19</date><risdate>2020</risdate><abstract>To provide an information processing apparatus configured to effectively increase the speed of learning using DNN regardless of learning methods, and an information processing method.SOLUTION: An information processing apparatus includes a learning unit which executes learning using a neural network. The learning unit dynamically modifies a value of batch size during learning on the basis of a gap value with an ideal state relating to learning output by the neural network. A batch size modifying unit dynamically modifies a value of batch size during the learning, which uses a neural network, on the basis of a gap value with an ideal state relating to learning output by the neural network.SELECTED DRAWING: Figure 3 【課題】DNNによる学習を学習手法に依らず効果的に高速化する情報処理装置および情報処理方法を提供する。【解決手段】情報処理装置は、ニューラルネットワークを用いた学習を行う学習部を備える。学習部は、ニューラルネットワークが出力する学習に係る理想状態とのギャップ値に基づいて、学習中にバッチサイズの値を動的に変更する。また、ニューラルネットワークを用いた学習を行うことを含み、バッチサイズ変更部は、ニューラルネットワークが出力する学習に係る理想状態とのギャップ値に基づいて、学習中にバッチサイズの値を動的に変更する。【選択図】図3</abstract><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier
ispartof
issn
language eng ; jpn
recordid cdi_epo_espacenet_JP2020042591A
source esp@cenet
subjects CALCULATING
COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
COMPUTING
COUNTING
PHYSICS
title INFORMATION PROCESSING APPARATUS AND INFORMATION PROCESSING METHOD
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-05T22%3A32%3A12IST&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=MIKAMI%20HIROAKI&rft.date=2020-03-19&rft_id=info:doi/&rft_dat=%3Cepo_EVB%3EJP2020042591A%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