REORDERING OF SPARSE DATA TO INDUCE SPATIAL LOCALITY FOR N-DIMENSIONAL SPARSE CONVOLUTIONAL NEURAL NETWORK PROCESSING
Exemplary embodiments maintain spatial locality of the data being processed by a sparse CNN. The spatial locality is maintained by reordering the data to preserve spatial locality. The reordering may be performed on data elements and on data for groups of co-located data elements referred to herein...
Saved in:
Main Authors | , , , |
---|---|
Format | Patent |
Language | English French German |
Published |
03.11.2021
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Abstract | Exemplary embodiments maintain spatial locality of the data being processed by a sparse CNN. The spatial locality is maintained by reordering the data to preserve spatial locality. The reordering may be performed on data elements and on data for groups of co-located data elements referred to herein as "chunks". Thus, the data may be reordered into chunks, where each chunk contains data for spatially co-located data elements, and in addition, chunks may be organized so that spatially located chunks are together. The use of chunks helps to reduce the need to re-fetch data during processing. Chunk sizes may be chosen based on the memory constraints of the processing logic (e.g., cache sizes). |
---|---|
AbstractList | Exemplary embodiments maintain spatial locality of the data being processed by a sparse CNN. The spatial locality is maintained by reordering the data to preserve spatial locality. The reordering may be performed on data elements and on data for groups of co-located data elements referred to herein as "chunks". Thus, the data may be reordered into chunks, where each chunk contains data for spatially co-located data elements, and in addition, chunks may be organized so that spatially located chunks are together. The use of chunks helps to reduce the need to re-fetch data during processing. Chunk sizes may be chosen based on the memory constraints of the processing logic (e.g., cache sizes). |
Author | Omer, Om Thyagharajan, Anirud Laddha, Prashant Subramoney, Sreenivas |
Author_xml | – fullname: Subramoney, Sreenivas – fullname: Omer, Om – fullname: Thyagharajan, Anirud – fullname: Laddha, Prashant |
BookMark | eNqNjcsKwjAQRbPQha9_mB8otFQXLkMy1WDMlEmquCpF4kraQvX_rY8PcHXg3Hu5czFpuzbOxJORWCMbtwMqwJeSPYKWQUIgME5XCt82GGnBkpLWhAsUxOASbY7ovCE3Rr-hInciW4WvdFjxB-FMfICSSaH349VSTG_NfYirHxcCCgxqn8S-q-PQN9fYxkeNZb5NN9k6lVn-R-UFC2065A |
ContentType | Patent |
DBID | EVB |
DatabaseName | esp@cenet |
DatabaseTitleList | |
Database_xml | – sequence: 1 dbid: EVB name: esp@cenet url: http://worldwide.espacenet.com/singleLineSearch?locale=en_EP sourceTypes: Open Access Repository |
DeliveryMethod | fulltext_linktorsrc |
Discipline | Medicine Chemistry Sciences Physics |
DocumentTitleAlternate | R�ORGANISATION DE DONN�ES RARES POUR INDUIRE UNE LOCALISATION SPATIALE POUR LE TRAITEMENT DE R�SEAU DE NEURONES ARTIFICIELS CONVOLUTIFS CREUX � N-DIMENSIONS NEUORDNUNG VON SPÄRLICHEN DATEN ZUM INDUZIEREN EINER RÄUMLICHEN LOKALITÄT ZUR VERARBEITUNG EINES N-DIMENSIONALEN, SPÄRLICHEN NEURONALEN FALTUNGSNETZES |
ExternalDocumentID | EP3905140A1 |
GroupedDBID | EVB |
ID | FETCH-epo_espacenet_EP3905140A13 |
IEDL.DBID | EVB |
IngestDate | Fri Jul 19 14:48:54 EDT 2024 |
IsOpenAccess | true |
IsPeerReviewed | false |
IsScholarly | false |
Language | English French German |
LinkModel | DirectLink |
MergedId | FETCHMERGED-epo_espacenet_EP3905140A13 |
Notes | Application Number: EP20200209681 |
OpenAccessLink | https://worldwide.espacenet.com/publicationDetails/biblio?FT=D&date=20211103&DB=EPODOC&CC=EP&NR=3905140A1 |
ParticipantIDs | epo_espacenet_EP3905140A1 |
PublicationCentury | 2000 |
PublicationDate | 20211103 |
PublicationDateYYYYMMDD | 2021-11-03 |
PublicationDate_xml | – month: 11 year: 2021 text: 20211103 day: 03 |
PublicationDecade | 2020 |
PublicationYear | 2021 |
RelatedCompanies | INTEL Corporation |
RelatedCompanies_xml | – name: INTEL Corporation |
Score | 3.357435 |
Snippet | Exemplary embodiments maintain spatial locality of the data being processed by a sparse CNN. The spatial locality is maintained by reordering the data to... |
SourceID | epo |
SourceType | Open Access Repository |
SubjectTerms | CALCULATING COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS COMPUTING COUNTING PHYSICS |
Title | REORDERING OF SPARSE DATA TO INDUCE SPATIAL LOCALITY FOR N-DIMENSIONAL SPARSE CONVOLUTIONAL NEURAL NETWORK PROCESSING |
URI | https://worldwide.espacenet.com/publicationDetails/biblio?FT=D&date=20211103&DB=EPODOC&locale=&CC=EP&NR=3905140A1 |
hasFullText | 1 |
inHoldings | 1 |
isFullTextHit | |
isPrint | |
link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwfV07T8MwED5VvDcoIMpLHlC2iJJHS4YKpY5DA6kdpUkpU5Wn1CWtaCr-PrZpCwtMts6yJZ91Z3-27zuAO0287OqlplqpZnGAknTVhB9r1TzTLKP7WBaajEob0s4gNl4m5qQBs00sjOQJ_ZTkiNyiMm7vtfTXi59LLEf-rVzepzMumj-5Uc9R1uiYo5mHtq44_R4JmMOwgjGvKTTs6YKHymjbHCjtilO0oNkn474ISln83lHcY9gL-GBVfQKNomrCId4kXmvCwXD93t2EfflBM1ty4doIl6ewCglHbpILCjEXjQI7HBHk2JGNIoY86sSYCGnk2T7yGbZ9L3pHHO0hqjrekNCR5L_ddMSMjpkff5PiIkriUBbRGwtfURAyLBwufT4D5JIID1Q-k-lWa1MSbOesn8NONa-KC0CmmXcyS0tFjmG-MnlamlpeZEnHLBOjbSQtaP05zOU_bVdwJNQvo_T0a9ipP1bFDd-u6_RWKvoLmNeQEA |
link.rule.ids | 230,309,786,891,25594,76906 |
linkProvider | European Patent Office |
linkToHtml | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwfV3NT8IwFG8IfuBNUSN-9mB2W8R9gDsQM7oik61dRod4IvtMuAwiI_77thXQi57avKZN-pr3Xn9t368A3GviZlcvNNVKNIsDlLirxnxbq2apZhndpyLXZFaaTzrDyHidmtMamG9zYSRP6KckR-QWlXJ7r6S_Xv4cYjnybeXqIZlz0eJ5wHqOskHHHM08tnXF6fdwQB2KFIR4TSFhTxc8VEbb5kBpryvIecXOadIXSSnL3xFlcAz2Az5YWZ2AWl42QQNtP15rgkN_c9_dBAfygWa64sKNEa5OwTrEHLlJLihIB3Ac2OEYQ8dmNmQUusSJEBZS5toe9CiyPZe9Q472IFEd18dkLPlvtx0RJRPqRd-kuJDgKJQFe6PhCAYhRcLhkpczAAeYoaHKZzLbaW2Gg92c9XNQLxdlfgGgaWad1NIS8ccwX5ksKUwty9O4Yxax0TbiFmj9OczlP213oDFkvjfzXDK6AkdiKWTGnn4N6tXHOr_hobtKbqXSvwB_mJL9 |
openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Apatent&rft.title=REORDERING+OF+SPARSE+DATA+TO+INDUCE+SPATIAL+LOCALITY+FOR+N-DIMENSIONAL+SPARSE+CONVOLUTIONAL+NEURAL+NETWORK+PROCESSING&rft.inventor=Subramoney%2C+Sreenivas&rft.inventor=Omer%2C+Om&rft.inventor=Thyagharajan%2C+Anirud&rft.inventor=Laddha%2C+Prashant&rft.date=2021-11-03&rft.externalDBID=A1&rft.externalDocID=EP3905140A1 |