DEEP NEURAL NETWORK ACCELERATING DEVICE AND OPERATION METHOD THEREOF
A deep neural network learning acceleration device according to an embodiment of the present application comprises: a calculation part that sequentially performs first and second calculations for a plurality of input data of a subset, according to a mini-batch gradient descent method; a determinatio...
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Patent |
| Sprache: | eng ; kor |
| 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 | PARK JONG SUN KIM GEON HO JO JOONG HO SHIN DONG YEOB |
| description | A deep neural network learning acceleration device according to an embodiment of the present application comprises: a calculation part that sequentially performs first and second calculations for a plurality of input data of a subset, according to a mini-batch gradient descent method; a determination part that determines each of the plurality of input data as one among the skip data or the learning data, based on a reliability matrix obtained according to the first calculation; and a control part that controls the calculation part to skip the second calculation for the skip data. Therefore, the present invention is capable of reducing energy consumption.
본 출원의 실시예에 따른 심층 신경망 학습 가속 장치는 미니-배치 경사 하강법에 따라, 서브 세트의 복수의 입력 데이터들에 대한 제1 및 제2 연산을 순차적으로 수행하는 연산부, 상기 제1 연산에 따라 획득되는 신뢰도 행렬에 기초하여, 상기 복수의 입력 데이터들 각각을 스킵 데이터와 학습 데이터 중 어느 하나로 판단하는 판단부 및 상기 스킵 데이터에 대한 상기 제2 연산을 스킵시키도록 상기 연산부를 제어하는 제어부를 포함한다. |
| format | Patent |
| fullrecord | <record><control><sourceid>epo_EVB</sourceid><recordid>TN_cdi_epo_espacenet_KR20220010383A</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>KR20220010383A</sourcerecordid><originalsourceid>FETCH-epo_espacenet_KR20220010383A3</originalsourceid><addsrcrecordid>eNrjZHBxcXUNUPBzDQ1y9AFSIeH-Qd4Kjs7Orj6uQY4hnn7uCi6uYZ7OrgqOfi4K_gFgQX8_BV_XEA9_F4UQD9cgV383HgbWtMSc4lReKM3NoOzmGuLsoZtakB-fWlyQmJyal1oS7x1kZGBkZGBgaGBsYexoTJwqAO_DLGQ</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>patent</recordtype></control><display><type>patent</type><title>DEEP NEURAL NETWORK ACCELERATING DEVICE AND OPERATION METHOD THEREOF</title><source>esp@cenet</source><creator>PARK JONG SUN ; KIM GEON HO ; JO JOONG HO ; SHIN DONG YEOB</creator><creatorcontrib>PARK JONG SUN ; KIM GEON HO ; JO JOONG HO ; SHIN DONG YEOB</creatorcontrib><description>A deep neural network learning acceleration device according to an embodiment of the present application comprises: a calculation part that sequentially performs first and second calculations for a plurality of input data of a subset, according to a mini-batch gradient descent method; a determination part that determines each of the plurality of input data as one among the skip data or the learning data, based on a reliability matrix obtained according to the first calculation; and a control part that controls the calculation part to skip the second calculation for the skip data. Therefore, the present invention is capable of reducing energy consumption.
본 출원의 실시예에 따른 심층 신경망 학습 가속 장치는 미니-배치 경사 하강법에 따라, 서브 세트의 복수의 입력 데이터들에 대한 제1 및 제2 연산을 순차적으로 수행하는 연산부, 상기 제1 연산에 따라 획득되는 신뢰도 행렬에 기초하여, 상기 복수의 입력 데이터들 각각을 스킵 데이터와 학습 데이터 중 어느 하나로 판단하는 판단부 및 상기 스킵 데이터에 대한 상기 제2 연산을 스킵시키도록 상기 연산부를 제어하는 제어부를 포함한다.</description><language>eng ; kor</language><subject>CALCULATING ; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS ; COMPUTING ; COUNTING ; PHYSICS</subject><creationdate>2022</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&date=20220125&DB=EPODOC&CC=KR&NR=20220010383A$$EHTML$$P50$$Gepo$$Hfree_for_read</linktohtml><link.rule.ids>230,308,777,882,25545,76296</link.rule.ids><linktorsrc>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&date=20220125&DB=EPODOC&CC=KR&NR=20220010383A$$EView_record_in_European_Patent_Office$$FView_record_in_$$GEuropean_Patent_Office$$Hfree_for_read</linktorsrc></links><search><creatorcontrib>PARK JONG SUN</creatorcontrib><creatorcontrib>KIM GEON HO</creatorcontrib><creatorcontrib>JO JOONG HO</creatorcontrib><creatorcontrib>SHIN DONG YEOB</creatorcontrib><title>DEEP NEURAL NETWORK ACCELERATING DEVICE AND OPERATION METHOD THEREOF</title><description>A deep neural network learning acceleration device according to an embodiment of the present application comprises: a calculation part that sequentially performs first and second calculations for a plurality of input data of a subset, according to a mini-batch gradient descent method; a determination part that determines each of the plurality of input data as one among the skip data or the learning data, based on a reliability matrix obtained according to the first calculation; and a control part that controls the calculation part to skip the second calculation for the skip data. Therefore, the present invention is capable of reducing energy consumption.
본 출원의 실시예에 따른 심층 신경망 학습 가속 장치는 미니-배치 경사 하강법에 따라, 서브 세트의 복수의 입력 데이터들에 대한 제1 및 제2 연산을 순차적으로 수행하는 연산부, 상기 제1 연산에 따라 획득되는 신뢰도 행렬에 기초하여, 상기 복수의 입력 데이터들 각각을 스킵 데이터와 학습 데이터 중 어느 하나로 판단하는 판단부 및 상기 스킵 데이터에 대한 상기 제2 연산을 스킵시키도록 상기 연산부를 제어하는 제어부를 포함한다.</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>2022</creationdate><recordtype>patent</recordtype><sourceid>EVB</sourceid><recordid>eNrjZHBxcXUNUPBzDQ1y9AFSIeH-Qd4Kjs7Orj6uQY4hnn7uCi6uYZ7OrgqOfi4K_gFgQX8_BV_XEA9_F4UQD9cgV383HgbWtMSc4lReKM3NoOzmGuLsoZtakB-fWlyQmJyal1oS7x1kZGBkZGBgaGBsYexoTJwqAO_DLGQ</recordid><startdate>20220125</startdate><enddate>20220125</enddate><creator>PARK JONG SUN</creator><creator>KIM GEON HO</creator><creator>JO JOONG HO</creator><creator>SHIN DONG YEOB</creator><scope>EVB</scope></search><sort><creationdate>20220125</creationdate><title>DEEP NEURAL NETWORK ACCELERATING DEVICE AND OPERATION METHOD THEREOF</title><author>PARK JONG SUN ; KIM GEON HO ; JO JOONG HO ; SHIN DONG YEOB</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-epo_espacenet_KR20220010383A3</frbrgroupid><rsrctype>patents</rsrctype><prefilter>patents</prefilter><language>eng ; kor</language><creationdate>2022</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>PARK JONG SUN</creatorcontrib><creatorcontrib>KIM GEON HO</creatorcontrib><creatorcontrib>JO JOONG HO</creatorcontrib><creatorcontrib>SHIN DONG YEOB</creatorcontrib><collection>esp@cenet</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>PARK JONG SUN</au><au>KIM GEON HO</au><au>JO JOONG HO</au><au>SHIN DONG YEOB</au><format>patent</format><genre>patent</genre><ristype>GEN</ristype><title>DEEP NEURAL NETWORK ACCELERATING DEVICE AND OPERATION METHOD THEREOF</title><date>2022-01-25</date><risdate>2022</risdate><abstract>A deep neural network learning acceleration device according to an embodiment of the present application comprises: a calculation part that sequentially performs first and second calculations for a plurality of input data of a subset, according to a mini-batch gradient descent method; a determination part that determines each of the plurality of input data as one among the skip data or the learning data, based on a reliability matrix obtained according to the first calculation; and a control part that controls the calculation part to skip the second calculation for the skip data. Therefore, the present invention is capable of reducing energy consumption.
본 출원의 실시예에 따른 심층 신경망 학습 가속 장치는 미니-배치 경사 하강법에 따라, 서브 세트의 복수의 입력 데이터들에 대한 제1 및 제2 연산을 순차적으로 수행하는 연산부, 상기 제1 연산에 따라 획득되는 신뢰도 행렬에 기초하여, 상기 복수의 입력 데이터들 각각을 스킵 데이터와 학습 데이터 중 어느 하나로 판단하는 판단부 및 상기 스킵 데이터에 대한 상기 제2 연산을 스킵시키도록 상기 연산부를 제어하는 제어부를 포함한다.</abstract><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | |
| ispartof | |
| issn | |
| language | eng ; kor |
| recordid | cdi_epo_espacenet_KR20220010383A |
| source | esp@cenet |
| subjects | CALCULATING COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS COMPUTING COUNTING PHYSICS |
| title | DEEP NEURAL NETWORK ACCELERATING DEVICE AND OPERATION METHOD THEREOF |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-17T14%3A11%3A20IST&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=PARK%20JONG%20SUN&rft.date=2022-01-25&rft_id=info:doi/&rft_dat=%3Cepo_EVB%3EKR20220010383A%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 |