Accelerating Sparse Deep Neural Networks Article Swipe
Asit Mishra
,
Jorge Albericio Latorre
,
Jeff Pool
,
Darko Stošić
,
Dušan Stošić
,
Ganesh Venkatesh
,
Chong Yu
,
Paulius Micikevicius
·
YOU?
·
· 2021
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2104.08378
YOU?
·
· 2021
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2104.08378
As neural network model sizes have dramatically increased, so has the interest in various techniques to reduce their parameter counts and accelerate their execution. An active area of research in this field is sparsity - encouraging zero values in parameters that can then be discarded from storage or computations. While most research focuses on high levels of sparsity, there are challenges in universally maintaining model accuracy as well as achieving significant speedups over modern matrix-math hardware. To make sparsity adoption practical, the NVIDIA Ampere GPU architecture introduces sparsity support in its matrix-math units, Tensor Cores. We present the design and behavior of Sparse Tensor Cores, which exploit a 2:4 (50%) sparsity pattern that leads to twice the math throughput of dense matrix units. We also describe a simple workflow for training networks that both satisfy 2:4 sparsity pattern requirements and maintain accuracy, verifying it on a wide range of common tasks and model architectures. This workflow makes it easy to prepare accurate models for efficient deployment on Sparse Tensor Cores.
Related Topics
Concepts
Computer science
Workflow
Tensor (intrinsic definition)
Exploit
Artificial neural network
Sparse matrix
Computation
Matrix (chemical analysis)
Simple (philosophy)
Software deployment
Range (aeronautics)
Throughput
Field (mathematics)
Computational science
Parallel computing
Computer engineering
Algorithm
Artificial intelligence
Mathematics
Wireless
Epistemology
Pure mathematics
Gaussian
Philosophy
Operating system
Physics
Materials science
Computer security
Telecommunications
Database
Quantum mechanics
Composite material
Metadata
- Type
- preprint
- Language
- en
- Landing Page
- http://arxiv.org/abs/2104.08378
- https://arxiv.org/pdf/2104.08378
- OA Status
- green
- Cited By
- 14
- References
- 52
- Related Works
- 20
- OpenAlex ID
- https://openalex.org/W3156528192
All OpenAlex metadata
Raw OpenAlex JSON
- OpenAlex ID
-
https://openalex.org/W3156528192Canonical identifier for this work in OpenAlex
- DOI
-
https://doi.org/10.48550/arxiv.2104.08378Digital Object Identifier
- Title
-
Accelerating Sparse Deep Neural NetworksWork title
- Type
-
preprintOpenAlex work type
- Language
-
enPrimary language
- Publication year
-
2021Year of publication
- Publication date
-
2021-04-16Full publication date if available
- Authors
-
Asit Mishra, Jorge Albericio Latorre, Jeff Pool, Darko Stošić, Dušan Stošić, Ganesh Venkatesh, Chong Yu, Paulius MicikeviciusList of authors in order
- Landing page
-
https://arxiv.org/abs/2104.08378Publisher landing page
- PDF URL
-
https://arxiv.org/pdf/2104.08378Direct link to full text PDF
- Open access
-
YesWhether a free full text is available
- OA status
-
greenOpen access status per OpenAlex
- OA URL
-
https://arxiv.org/pdf/2104.08378Direct OA link when available
- Concepts
-
Computer science, Workflow, Tensor (intrinsic definition), Exploit, Artificial neural network, Sparse matrix, Computation, Matrix (chemical analysis), Simple (philosophy), Software deployment, Range (aeronautics), Throughput, Field (mathematics), Computational science, Parallel computing, Computer engineering, Algorithm, Artificial intelligence, Mathematics, Wireless, Epistemology, Pure mathematics, Gaussian, Philosophy, Operating system, Physics, Materials science, Computer security, Telecommunications, Database, Quantum mechanics, Composite materialTop concepts (fields/topics) attached by OpenAlex
- Cited by
-
14Total citation count in OpenAlex
- Citations by year (recent)
-
2025: 1, 2024: 1, 2023: 1, 2022: 1, 2021: 10Per-year citation counts (last 5 years)
- References (count)
-
52Number of works referenced by this work
- Related works (count)
-
20Other works algorithmically related by OpenAlex
Full payload
| id | https://openalex.org/W3156528192 |
|---|---|
| doi | https://doi.org/10.48550/arxiv.2104.08378 |
| ids.doi | https://doi.org/10.48550/arxiv.2104.08378 |
| ids.mag | 3156528192 |
| ids.openalex | https://openalex.org/W3156528192 |
| fwci | |
| type | preprint |
| title | Accelerating Sparse Deep Neural Networks |
| biblio.issue | |
| biblio.volume | |
| biblio.last_page | |
| biblio.first_page | |
| topics[0].id | https://openalex.org/T10036 |
| topics[0].field.id | https://openalex.org/fields/17 |
| topics[0].field.display_name | Computer Science |
| topics[0].score | 0.9983999729156494 |
| topics[0].domain.id | https://openalex.org/domains/3 |
| topics[0].domain.display_name | Physical Sciences |
| topics[0].subfield.id | https://openalex.org/subfields/1707 |
| topics[0].subfield.display_name | Computer Vision and Pattern Recognition |
| topics[0].display_name | Advanced Neural Network Applications |
| topics[1].id | https://openalex.org/T12303 |
| topics[1].field.id | https://openalex.org/fields/26 |
| topics[1].field.display_name | Mathematics |
| topics[1].score | 0.9965000152587891 |
| topics[1].domain.id | https://openalex.org/domains/3 |
| topics[1].domain.display_name | Physical Sciences |
| topics[1].subfield.id | https://openalex.org/subfields/2605 |
| topics[1].subfield.display_name | Computational Mathematics |
| topics[1].display_name | Tensor decomposition and applications |
| topics[2].id | https://openalex.org/T10054 |
| topics[2].field.id | https://openalex.org/fields/17 |
| topics[2].field.display_name | Computer Science |
| topics[2].score | 0.9933000206947327 |
| topics[2].domain.id | https://openalex.org/domains/3 |
| topics[2].domain.display_name | Physical Sciences |
| topics[2].subfield.id | https://openalex.org/subfields/1708 |
| topics[2].subfield.display_name | Hardware and Architecture |
| topics[2].display_name | Parallel Computing and Optimization Techniques |
| is_xpac | False |
| apc_list | |
| apc_paid | |
| concepts[0].id | https://openalex.org/C41008148 |
| concepts[0].level | 0 |
| concepts[0].score | 0.7751962542533875 |
| concepts[0].wikidata | https://www.wikidata.org/wiki/Q21198 |
| concepts[0].display_name | Computer science |
| concepts[1].id | https://openalex.org/C177212765 |
| concepts[1].level | 2 |
| concepts[1].score | 0.7535868883132935 |
| concepts[1].wikidata | https://www.wikidata.org/wiki/Q627335 |
| concepts[1].display_name | Workflow |
| concepts[2].id | https://openalex.org/C155281189 |
| concepts[2].level | 2 |
| concepts[2].score | 0.60260409116745 |
| concepts[2].wikidata | https://www.wikidata.org/wiki/Q3518150 |
| concepts[2].display_name | Tensor (intrinsic definition) |
| concepts[3].id | https://openalex.org/C165696696 |
| concepts[3].level | 2 |
| concepts[3].score | 0.5791079998016357 |
| concepts[3].wikidata | https://www.wikidata.org/wiki/Q11287 |
| concepts[3].display_name | Exploit |
| concepts[4].id | https://openalex.org/C50644808 |
| concepts[4].level | 2 |
| concepts[4].score | 0.5432116389274597 |
| concepts[4].wikidata | https://www.wikidata.org/wiki/Q192776 |
| concepts[4].display_name | Artificial neural network |
| concepts[5].id | https://openalex.org/C56372850 |
| concepts[5].level | 3 |
| concepts[5].score | 0.5387692451477051 |
| concepts[5].wikidata | https://www.wikidata.org/wiki/Q1050404 |
| concepts[5].display_name | Sparse matrix |
| concepts[6].id | https://openalex.org/C45374587 |
| concepts[6].level | 2 |
| concepts[6].score | 0.5151699781417847 |
| concepts[6].wikidata | https://www.wikidata.org/wiki/Q12525525 |
| concepts[6].display_name | Computation |
| concepts[7].id | https://openalex.org/C106487976 |
| concepts[7].level | 2 |
| concepts[7].score | 0.5132613182067871 |
| concepts[7].wikidata | https://www.wikidata.org/wiki/Q685816 |
| concepts[7].display_name | Matrix (chemical analysis) |
| concepts[8].id | https://openalex.org/C2780586882 |
| concepts[8].level | 2 |
| concepts[8].score | 0.49130645394325256 |
| concepts[8].wikidata | https://www.wikidata.org/wiki/Q7520643 |
| concepts[8].display_name | Simple (philosophy) |
| concepts[9].id | https://openalex.org/C105339364 |
| concepts[9].level | 2 |
| concepts[9].score | 0.48567402362823486 |
| concepts[9].wikidata | https://www.wikidata.org/wiki/Q2297740 |
| concepts[9].display_name | Software deployment |
| concepts[10].id | https://openalex.org/C204323151 |
| concepts[10].level | 2 |
| concepts[10].score | 0.465629518032074 |
| concepts[10].wikidata | https://www.wikidata.org/wiki/Q905424 |
| concepts[10].display_name | Range (aeronautics) |
| concepts[11].id | https://openalex.org/C157764524 |
| concepts[11].level | 3 |
| concepts[11].score | 0.4483739137649536 |
| concepts[11].wikidata | https://www.wikidata.org/wiki/Q1383412 |
| concepts[11].display_name | Throughput |
| concepts[12].id | https://openalex.org/C9652623 |
| concepts[12].level | 2 |
| concepts[12].score | 0.4365934133529663 |
| concepts[12].wikidata | https://www.wikidata.org/wiki/Q190109 |
| concepts[12].display_name | Field (mathematics) |
| concepts[13].id | https://openalex.org/C459310 |
| concepts[13].level | 1 |
| concepts[13].score | 0.4242706000804901 |
| concepts[13].wikidata | https://www.wikidata.org/wiki/Q117801 |
| concepts[13].display_name | Computational science |
| concepts[14].id | https://openalex.org/C173608175 |
| concepts[14].level | 1 |
| concepts[14].score | 0.4094640612602234 |
| concepts[14].wikidata | https://www.wikidata.org/wiki/Q232661 |
| concepts[14].display_name | Parallel computing |
| concepts[15].id | https://openalex.org/C113775141 |
| concepts[15].level | 1 |
| concepts[15].score | 0.3535151481628418 |
| concepts[15].wikidata | https://www.wikidata.org/wiki/Q428691 |
| concepts[15].display_name | Computer engineering |
| concepts[16].id | https://openalex.org/C11413529 |
| concepts[16].level | 1 |
| concepts[16].score | 0.3525136411190033 |
| concepts[16].wikidata | https://www.wikidata.org/wiki/Q8366 |
| concepts[16].display_name | Algorithm |
| concepts[17].id | https://openalex.org/C154945302 |
| concepts[17].level | 1 |
| concepts[17].score | 0.31614676117897034 |
| concepts[17].wikidata | https://www.wikidata.org/wiki/Q11660 |
| concepts[17].display_name | Artificial intelligence |
| concepts[18].id | https://openalex.org/C33923547 |
| concepts[18].level | 0 |
| concepts[18].score | 0.11064726114273071 |
| concepts[18].wikidata | https://www.wikidata.org/wiki/Q395 |
| concepts[18].display_name | Mathematics |
| concepts[19].id | https://openalex.org/C555944384 |
| concepts[19].level | 2 |
| concepts[19].score | 0.09592649340629578 |
| concepts[19].wikidata | https://www.wikidata.org/wiki/Q249 |
| concepts[19].display_name | Wireless |
| concepts[20].id | https://openalex.org/C111472728 |
| concepts[20].level | 1 |
| concepts[20].score | 0.0 |
| concepts[20].wikidata | https://www.wikidata.org/wiki/Q9471 |
| concepts[20].display_name | Epistemology |
| concepts[21].id | https://openalex.org/C202444582 |
| concepts[21].level | 1 |
| concepts[21].score | 0.0 |
| concepts[21].wikidata | https://www.wikidata.org/wiki/Q837863 |
| concepts[21].display_name | Pure mathematics |
| concepts[22].id | https://openalex.org/C163716315 |
| concepts[22].level | 2 |
| concepts[22].score | 0.0 |
| concepts[22].wikidata | https://www.wikidata.org/wiki/Q901177 |
| concepts[22].display_name | Gaussian |
| concepts[23].id | https://openalex.org/C138885662 |
| concepts[23].level | 0 |
| concepts[23].score | 0.0 |
| concepts[23].wikidata | https://www.wikidata.org/wiki/Q5891 |
| concepts[23].display_name | Philosophy |
| concepts[24].id | https://openalex.org/C111919701 |
| concepts[24].level | 1 |
| concepts[24].score | 0.0 |
| concepts[24].wikidata | https://www.wikidata.org/wiki/Q9135 |
| concepts[24].display_name | Operating system |
| concepts[25].id | https://openalex.org/C121332964 |
| concepts[25].level | 0 |
| concepts[25].score | 0.0 |
| concepts[25].wikidata | https://www.wikidata.org/wiki/Q413 |
| concepts[25].display_name | Physics |
| concepts[26].id | https://openalex.org/C192562407 |
| concepts[26].level | 0 |
| concepts[26].score | 0.0 |
| concepts[26].wikidata | https://www.wikidata.org/wiki/Q228736 |
| concepts[26].display_name | Materials science |
| concepts[27].id | https://openalex.org/C38652104 |
| concepts[27].level | 1 |
| concepts[27].score | 0.0 |
| concepts[27].wikidata | https://www.wikidata.org/wiki/Q3510521 |
| concepts[27].display_name | Computer security |
| concepts[28].id | https://openalex.org/C76155785 |
| concepts[28].level | 1 |
| concepts[28].score | 0.0 |
| concepts[28].wikidata | https://www.wikidata.org/wiki/Q418 |
| concepts[28].display_name | Telecommunications |
| concepts[29].id | https://openalex.org/C77088390 |
| concepts[29].level | 1 |
| concepts[29].score | 0.0 |
| concepts[29].wikidata | https://www.wikidata.org/wiki/Q8513 |
| concepts[29].display_name | Database |
| concepts[30].id | https://openalex.org/C62520636 |
| concepts[30].level | 1 |
| concepts[30].score | 0.0 |
| concepts[30].wikidata | https://www.wikidata.org/wiki/Q944 |
| concepts[30].display_name | Quantum mechanics |
| concepts[31].id | https://openalex.org/C159985019 |
| concepts[31].level | 1 |
| concepts[31].score | 0.0 |
| concepts[31].wikidata | https://www.wikidata.org/wiki/Q181790 |
| concepts[31].display_name | Composite material |
| keywords[0].id | https://openalex.org/keywords/computer-science |
| keywords[0].score | 0.7751962542533875 |
| keywords[0].display_name | Computer science |
| keywords[1].id | https://openalex.org/keywords/workflow |
| keywords[1].score | 0.7535868883132935 |
| keywords[1].display_name | Workflow |
| keywords[2].id | https://openalex.org/keywords/tensor |
| keywords[2].score | 0.60260409116745 |
| keywords[2].display_name | Tensor (intrinsic definition) |
| keywords[3].id | https://openalex.org/keywords/exploit |
| keywords[3].score | 0.5791079998016357 |
| keywords[3].display_name | Exploit |
| keywords[4].id | https://openalex.org/keywords/artificial-neural-network |
| keywords[4].score | 0.5432116389274597 |
| keywords[4].display_name | Artificial neural network |
| keywords[5].id | https://openalex.org/keywords/sparse-matrix |
| keywords[5].score | 0.5387692451477051 |
| keywords[5].display_name | Sparse matrix |
| keywords[6].id | https://openalex.org/keywords/computation |
| keywords[6].score | 0.5151699781417847 |
| keywords[6].display_name | Computation |
| keywords[7].id | https://openalex.org/keywords/matrix |
| keywords[7].score | 0.5132613182067871 |
| keywords[7].display_name | Matrix (chemical analysis) |
| keywords[8].id | https://openalex.org/keywords/simple |
| keywords[8].score | 0.49130645394325256 |
| keywords[8].display_name | Simple (philosophy) |
| keywords[9].id | https://openalex.org/keywords/software-deployment |
| keywords[9].score | 0.48567402362823486 |
| keywords[9].display_name | Software deployment |
| keywords[10].id | https://openalex.org/keywords/range |
| keywords[10].score | 0.465629518032074 |
| keywords[10].display_name | Range (aeronautics) |
| keywords[11].id | https://openalex.org/keywords/throughput |
| keywords[11].score | 0.4483739137649536 |
| keywords[11].display_name | Throughput |
| keywords[12].id | https://openalex.org/keywords/field |
| keywords[12].score | 0.4365934133529663 |
| keywords[12].display_name | Field (mathematics) |
| keywords[13].id | https://openalex.org/keywords/computational-science |
| keywords[13].score | 0.4242706000804901 |
| keywords[13].display_name | Computational science |
| keywords[14].id | https://openalex.org/keywords/parallel-computing |
| keywords[14].score | 0.4094640612602234 |
| keywords[14].display_name | Parallel computing |
| keywords[15].id | https://openalex.org/keywords/computer-engineering |
| keywords[15].score | 0.3535151481628418 |
| keywords[15].display_name | Computer engineering |
| keywords[16].id | https://openalex.org/keywords/algorithm |
| keywords[16].score | 0.3525136411190033 |
| keywords[16].display_name | Algorithm |
| keywords[17].id | https://openalex.org/keywords/artificial-intelligence |
| keywords[17].score | 0.31614676117897034 |
| keywords[17].display_name | Artificial intelligence |
| keywords[18].id | https://openalex.org/keywords/mathematics |
| keywords[18].score | 0.11064726114273071 |
| keywords[18].display_name | Mathematics |
| keywords[19].id | https://openalex.org/keywords/wireless |
| keywords[19].score | 0.09592649340629578 |
| keywords[19].display_name | Wireless |
| language | en |
| locations[0].id | pmh:oai:arXiv.org:2104.08378 |
| locations[0].is_oa | True |
| locations[0].source.id | https://openalex.org/S4306400194 |
| locations[0].source.issn | |
| locations[0].source.type | repository |
| locations[0].source.is_oa | True |
| locations[0].source.issn_l | |
| locations[0].source.is_core | False |
| locations[0].source.is_in_doaj | False |
| locations[0].source.display_name | arXiv (Cornell University) |
| locations[0].source.host_organization | https://openalex.org/I205783295 |
| locations[0].source.host_organization_name | Cornell University |
| locations[0].source.host_organization_lineage | https://openalex.org/I205783295 |
| locations[0].license | |
| locations[0].pdf_url | https://arxiv.org/pdf/2104.08378 |
| locations[0].version | submittedVersion |
| locations[0].raw_type | text |
| locations[0].license_id | |
| locations[0].is_accepted | False |
| locations[0].is_published | False |
| locations[0].raw_source_name | |
| locations[0].landing_page_url | http://arxiv.org/abs/2104.08378 |
| locations[1].id | mag:3156528192 |
| locations[1].is_oa | True |
| locations[1].source.id | https://openalex.org/S4306400194 |
| locations[1].source.issn | |
| locations[1].source.type | repository |
| locations[1].source.is_oa | True |
| locations[1].source.issn_l | |
| locations[1].source.is_core | False |
| locations[1].source.is_in_doaj | False |
| locations[1].source.display_name | arXiv (Cornell University) |
| locations[1].source.host_organization | https://openalex.org/I205783295 |
| locations[1].source.host_organization_name | Cornell University |
| locations[1].source.host_organization_lineage | https://openalex.org/I205783295 |
| locations[1].license | |
| locations[1].pdf_url | |
| locations[1].version | submittedVersion |
| locations[1].raw_type | |
| locations[1].license_id | |
| locations[1].is_accepted | False |
| locations[1].is_published | False |
| locations[1].raw_source_name | arXiv (Cornell University) |
| locations[1].landing_page_url | http://export.arxiv.org/pdf/2104.08378 |
| locations[2].id | doi:10.48550/arxiv.2104.08378 |
| locations[2].is_oa | True |
| locations[2].source.id | https://openalex.org/S4306400194 |
| locations[2].source.issn | |
| locations[2].source.type | repository |
| locations[2].source.is_oa | True |
| locations[2].source.issn_l | |
| locations[2].source.is_core | False |
| locations[2].source.is_in_doaj | False |
| locations[2].source.display_name | arXiv (Cornell University) |
| locations[2].source.host_organization | https://openalex.org/I205783295 |
| locations[2].source.host_organization_name | Cornell University |
| locations[2].source.host_organization_lineage | https://openalex.org/I205783295 |
| locations[2].license | |
| locations[2].pdf_url | |
| locations[2].version | |
| locations[2].raw_type | article |
| locations[2].license_id | |
| locations[2].is_accepted | False |
| locations[2].is_published | |
| locations[2].raw_source_name | |
| locations[2].landing_page_url | https://doi.org/10.48550/arxiv.2104.08378 |
| indexed_in | arxiv, datacite |
| authorships[0].author.id | https://openalex.org/A5101912070 |
| authorships[0].author.orcid | https://orcid.org/0000-0001-6489-6895 |
| authorships[0].author.display_name | Asit Mishra |
| authorships[0].countries | GB |
| authorships[0].affiliations[0].institution_ids | https://openalex.org/I1304085615 |
| authorships[0].affiliations[0].raw_affiliation_string | nVidia |
| authorships[0].institutions[0].id | https://openalex.org/I1304085615 |
| authorships[0].institutions[0].ror | https://ror.org/02kr42612 |
| authorships[0].institutions[0].type | company |
| authorships[0].institutions[0].lineage | https://openalex.org/I1304085615, https://openalex.org/I4210127875 |
| authorships[0].institutions[0].country_code | GB |
| authorships[0].institutions[0].display_name | Nvidia (United Kingdom) |
| authorships[0].author_position | first |
| authorships[0].raw_author_name | Asit K. Mishra |
| authorships[0].is_corresponding | False |
| authorships[0].raw_affiliation_strings | nVidia |
| authorships[1].author.id | https://openalex.org/A5004205517 |
| authorships[1].author.orcid | |
| authorships[1].author.display_name | Jorge Albericio Latorre |
| authorships[1].author_position | middle |
| authorships[1].raw_author_name | Jorge Albericio Latorre |
| authorships[1].is_corresponding | False |
| authorships[2].author.id | https://openalex.org/A5063722719 |
| authorships[2].author.orcid | |
| authorships[2].author.display_name | Jeff Pool |
| authorships[2].author_position | middle |
| authorships[2].raw_author_name | Jeff Pool |
| authorships[2].is_corresponding | False |
| authorships[3].author.id | https://openalex.org/A5026607041 |
| authorships[3].author.orcid | |
| authorships[3].author.display_name | Darko Stošić |
| authorships[3].author_position | middle |
| authorships[3].raw_author_name | Darko Stosic |
| authorships[3].is_corresponding | False |
| authorships[4].author.id | https://openalex.org/A5102721508 |
| authorships[4].author.orcid | https://orcid.org/0000-0003-0337-8511 |
| authorships[4].author.display_name | Dušan Stošić |
| authorships[4].author_position | middle |
| authorships[4].raw_author_name | Dusan Stosic |
| authorships[4].is_corresponding | False |
| authorships[5].author.id | https://openalex.org/A5104363400 |
| authorships[5].author.orcid | |
| authorships[5].author.display_name | Ganesh Venkatesh |
| authorships[5].author_position | middle |
| authorships[5].raw_author_name | Ganesh Venkatesh |
| authorships[5].is_corresponding | False |
| authorships[6].author.id | https://openalex.org/A5001459619 |
| authorships[6].author.orcid | https://orcid.org/0000-0002-6244-3486 |
| authorships[6].author.display_name | Chong Yu |
| authorships[6].author_position | middle |
| authorships[6].raw_author_name | Chong Yu |
| authorships[6].is_corresponding | False |
| authorships[7].author.id | https://openalex.org/A5072693649 |
| authorships[7].author.orcid | |
| authorships[7].author.display_name | Paulius Micikevicius |
| authorships[7].author_position | last |
| authorships[7].raw_author_name | Paulius Micikevicius |
| authorships[7].is_corresponding | False |
| has_content.pdf | False |
| has_content.grobid_xml | False |
| is_paratext | False |
| open_access.is_oa | True |
| open_access.oa_url | https://arxiv.org/pdf/2104.08378 |
| open_access.oa_status | green |
| open_access.any_repository_has_fulltext | False |
| created_date | 2025-10-10T00:00:00 |
| display_name | Accelerating Sparse Deep Neural Networks |
| has_fulltext | False |
| is_retracted | False |
| updated_date | 2025-11-06T06:51:31.235846 |
| primary_topic.id | https://openalex.org/T10036 |
| primary_topic.field.id | https://openalex.org/fields/17 |
| primary_topic.field.display_name | Computer Science |
| primary_topic.score | 0.9983999729156494 |
| primary_topic.domain.id | https://openalex.org/domains/3 |
| primary_topic.domain.display_name | Physical Sciences |
| primary_topic.subfield.id | https://openalex.org/subfields/1707 |
| primary_topic.subfield.display_name | Computer Vision and Pattern Recognition |
| primary_topic.display_name | Advanced Neural Network Applications |
| related_works | https://openalex.org/W2194775991, https://openalex.org/W2963674932, https://openalex.org/W3132616766, https://openalex.org/W3210792685, https://openalex.org/W2970971581, https://openalex.org/W2114766824, https://openalex.org/W2899244816, https://openalex.org/W2890498784, https://openalex.org/W3035839559, https://openalex.org/W3185569094, https://openalex.org/W3092640258, https://openalex.org/W2980186997, https://openalex.org/W3210117238, https://openalex.org/W3101555929, https://openalex.org/W2245317843, https://openalex.org/W3175408384, https://openalex.org/W2906007643, https://openalex.org/W2963813662, https://openalex.org/W2963341956, https://openalex.org/W2885392409 |
| cited_by_count | 14 |
| counts_by_year[0].year | 2025 |
| counts_by_year[0].cited_by_count | 1 |
| counts_by_year[1].year | 2024 |
| counts_by_year[1].cited_by_count | 1 |
| counts_by_year[2].year | 2023 |
| counts_by_year[2].cited_by_count | 1 |
| counts_by_year[3].year | 2022 |
| counts_by_year[3].cited_by_count | 1 |
| counts_by_year[4].year | 2021 |
| counts_by_year[4].cited_by_count | 10 |
| locations_count | 3 |
| best_oa_location.id | pmh:oai:arXiv.org:2104.08378 |
| best_oa_location.is_oa | True |
| best_oa_location.source.id | https://openalex.org/S4306400194 |
| best_oa_location.source.issn | |
| best_oa_location.source.type | repository |
| best_oa_location.source.is_oa | True |
| best_oa_location.source.issn_l | |
| best_oa_location.source.is_core | False |
| best_oa_location.source.is_in_doaj | False |
| best_oa_location.source.display_name | arXiv (Cornell University) |
| best_oa_location.source.host_organization | https://openalex.org/I205783295 |
| best_oa_location.source.host_organization_name | Cornell University |
| best_oa_location.source.host_organization_lineage | https://openalex.org/I205783295 |
| best_oa_location.license | |
| best_oa_location.pdf_url | https://arxiv.org/pdf/2104.08378 |
| best_oa_location.version | submittedVersion |
| best_oa_location.raw_type | text |
| best_oa_location.license_id | |
| best_oa_location.is_accepted | False |
| best_oa_location.is_published | False |
| best_oa_location.raw_source_name | |
| best_oa_location.landing_page_url | http://arxiv.org/abs/2104.08378 |
| primary_location.id | pmh:oai:arXiv.org:2104.08378 |
| primary_location.is_oa | True |
| primary_location.source.id | https://openalex.org/S4306400194 |
| primary_location.source.issn | |
| primary_location.source.type | repository |
| primary_location.source.is_oa | True |
| primary_location.source.issn_l | |
| primary_location.source.is_core | False |
| primary_location.source.is_in_doaj | False |
| primary_location.source.display_name | arXiv (Cornell University) |
| primary_location.source.host_organization | https://openalex.org/I205783295 |
| primary_location.source.host_organization_name | Cornell University |
| primary_location.source.host_organization_lineage | https://openalex.org/I205783295 |
| primary_location.license | |
| primary_location.pdf_url | https://arxiv.org/pdf/2104.08378 |
| primary_location.version | submittedVersion |
| primary_location.raw_type | text |
| primary_location.license_id | |
| primary_location.is_accepted | False |
| primary_location.is_published | False |
| primary_location.raw_source_name | |
| primary_location.landing_page_url | http://arxiv.org/abs/2104.08378 |
| publication_date | 2021-04-16 |
| publication_year | 2021 |
| referenced_works | https://openalex.org/W2619538244, https://openalex.org/W2963981733, https://openalex.org/W2951569836, https://openalex.org/W2914554687, https://openalex.org/W3098903812, https://openalex.org/W2764043458, https://openalex.org/W3132669550, https://openalex.org/W2962835968, https://openalex.org/W2904640605, https://openalex.org/W2194775991, https://openalex.org/W2986700214, https://openalex.org/W2964307104, https://openalex.org/W2943152387, https://openalex.org/W2799269579, https://openalex.org/W3098099022, https://openalex.org/W2792760996, https://openalex.org/W3017746288, https://openalex.org/W2901724498, https://openalex.org/W2707890836, https://openalex.org/W2955419283, https://openalex.org/W2963918968, https://openalex.org/W2883780447, https://openalex.org/W2899244816, https://openalex.org/W2900096133, https://openalex.org/W2933138175, https://openalex.org/W2963674932, https://openalex.org/W2963341956, https://openalex.org/W2737121650, https://openalex.org/W2990844796, https://openalex.org/W2276892413, https://openalex.org/W2767785892, https://openalex.org/W2427527485, https://openalex.org/W3127067080, https://openalex.org/W2509836288, https://openalex.org/W2955425717, https://openalex.org/W2911109671, https://openalex.org/W3002009013, https://openalex.org/W2798816912, https://openalex.org/W2796438033, https://openalex.org/W133229983, https://openalex.org/W3000181403, https://openalex.org/W2919347051, https://openalex.org/W2964254867, https://openalex.org/W2525778437, https://openalex.org/W2117539524, https://openalex.org/W2980186997, https://openalex.org/W2963247446, https://openalex.org/W2736953746, https://openalex.org/W2798275680, https://openalex.org/W2279098554, https://openalex.org/W2513419314, https://openalex.org/W2608554408 |
| referenced_works_count | 52 |
| abstract_inverted_index.- | 34 |
| abstract_inverted_index.a | 107, 126, 145 |
| abstract_inverted_index.An | 24 |
| abstract_inverted_index.As | 0 |
| abstract_inverted_index.To | 76 |
| abstract_inverted_index.We | 95, 123 |
| abstract_inverted_index.as | 66, 68 |
| abstract_inverted_index.be | 43 |
| abstract_inverted_index.in | 12, 29, 38, 61, 89 |
| abstract_inverted_index.is | 32 |
| abstract_inverted_index.it | 143, 157 |
| abstract_inverted_index.of | 27, 56, 101, 119, 148 |
| abstract_inverted_index.on | 53, 144, 166 |
| abstract_inverted_index.or | 47 |
| abstract_inverted_index.so | 8 |
| abstract_inverted_index.to | 15, 114, 159 |
| abstract_inverted_index.2:4 | 108, 135 |
| abstract_inverted_index.GPU | 84 |
| abstract_inverted_index.and | 20, 99, 139, 151 |
| abstract_inverted_index.are | 59 |
| abstract_inverted_index.can | 41 |
| abstract_inverted_index.for | 129, 163 |
| abstract_inverted_index.has | 9 |
| abstract_inverted_index.its | 90 |
| abstract_inverted_index.the | 10, 81, 97, 116 |
| abstract_inverted_index.This | 154 |
| abstract_inverted_index.also | 124 |
| abstract_inverted_index.area | 26 |
| abstract_inverted_index.both | 133 |
| abstract_inverted_index.easy | 158 |
| abstract_inverted_index.from | 45 |
| abstract_inverted_index.have | 5 |
| abstract_inverted_index.high | 54 |
| abstract_inverted_index.make | 77 |
| abstract_inverted_index.math | 117 |
| abstract_inverted_index.most | 50 |
| abstract_inverted_index.over | 72 |
| abstract_inverted_index.that | 40, 112, 132 |
| abstract_inverted_index.then | 42 |
| abstract_inverted_index.this | 30 |
| abstract_inverted_index.well | 67 |
| abstract_inverted_index.wide | 146 |
| abstract_inverted_index.zero | 36 |
| abstract_inverted_index.(50%) | 109 |
| abstract_inverted_index.While | 49 |
| abstract_inverted_index.dense | 120 |
| abstract_inverted_index.field | 31 |
| abstract_inverted_index.leads | 113 |
| abstract_inverted_index.makes | 156 |
| abstract_inverted_index.model | 3, 64, 152 |
| abstract_inverted_index.range | 147 |
| abstract_inverted_index.sizes | 4 |
| abstract_inverted_index.tasks | 150 |
| abstract_inverted_index.their | 17, 22 |
| abstract_inverted_index.there | 58 |
| abstract_inverted_index.twice | 115 |
| abstract_inverted_index.which | 105 |
| abstract_inverted_index.Ampere | 83 |
| abstract_inverted_index.Cores, | 104 |
| abstract_inverted_index.Cores. | 94, 169 |
| abstract_inverted_index.NVIDIA | 82 |
| abstract_inverted_index.Sparse | 102, 167 |
| abstract_inverted_index.Tensor | 93, 103, 168 |
| abstract_inverted_index.active | 25 |
| abstract_inverted_index.common | 149 |
| abstract_inverted_index.counts | 19 |
| abstract_inverted_index.design | 98 |
| abstract_inverted_index.levels | 55 |
| abstract_inverted_index.matrix | 121 |
| abstract_inverted_index.models | 162 |
| abstract_inverted_index.modern | 73 |
| abstract_inverted_index.neural | 1 |
| abstract_inverted_index.reduce | 16 |
| abstract_inverted_index.simple | 127 |
| abstract_inverted_index.units, | 92 |
| abstract_inverted_index.units. | 122 |
| abstract_inverted_index.values | 37 |
| abstract_inverted_index.exploit | 106 |
| abstract_inverted_index.focuses | 52 |
| abstract_inverted_index.network | 2 |
| abstract_inverted_index.pattern | 111, 137 |
| abstract_inverted_index.prepare | 160 |
| abstract_inverted_index.present | 96 |
| abstract_inverted_index.satisfy | 134 |
| abstract_inverted_index.storage | 46 |
| abstract_inverted_index.support | 88 |
| abstract_inverted_index.various | 13 |
| abstract_inverted_index.accuracy | 65 |
| abstract_inverted_index.accurate | 161 |
| abstract_inverted_index.adoption | 79 |
| abstract_inverted_index.behavior | 100 |
| abstract_inverted_index.describe | 125 |
| abstract_inverted_index.interest | 11 |
| abstract_inverted_index.maintain | 140 |
| abstract_inverted_index.networks | 131 |
| abstract_inverted_index.research | 28, 51 |
| abstract_inverted_index.sparsity | 33, 78, 87, 110, 136 |
| abstract_inverted_index.speedups | 71 |
| abstract_inverted_index.training | 130 |
| abstract_inverted_index.workflow | 128, 155 |
| abstract_inverted_index.accuracy, | 141 |
| abstract_inverted_index.achieving | 69 |
| abstract_inverted_index.discarded | 44 |
| abstract_inverted_index.efficient | 164 |
| abstract_inverted_index.hardware. | 75 |
| abstract_inverted_index.parameter | 18 |
| abstract_inverted_index.sparsity, | 57 |
| abstract_inverted_index.verifying | 142 |
| abstract_inverted_index.accelerate | 21 |
| abstract_inverted_index.challenges | 60 |
| abstract_inverted_index.deployment | 165 |
| abstract_inverted_index.execution. | 23 |
| abstract_inverted_index.increased, | 7 |
| abstract_inverted_index.introduces | 86 |
| abstract_inverted_index.parameters | 39 |
| abstract_inverted_index.practical, | 80 |
| abstract_inverted_index.techniques | 14 |
| abstract_inverted_index.throughput | 118 |
| abstract_inverted_index.encouraging | 35 |
| abstract_inverted_index.maintaining | 63 |
| abstract_inverted_index.matrix-math | 74, 91 |
| abstract_inverted_index.significant | 70 |
| abstract_inverted_index.universally | 62 |
| abstract_inverted_index.architecture | 85 |
| abstract_inverted_index.dramatically | 6 |
| abstract_inverted_index.requirements | 138 |
| abstract_inverted_index.computations. | 48 |
| abstract_inverted_index.architectures. | 153 |
| cited_by_percentile_year | |
| countries_distinct_count | 1 |
| institutions_distinct_count | 8 |
| citation_normalized_percentile |