Scalpel Article Swipe
Jiecao Yu
,
Andrew Lukefahr
,
David J. Palframan
,
Ganesh Dasika
,
Reetuparna Das
,
Scott Mahlke
·
YOU?
·
· 2017
· Open Access
·
· DOI: https://doi.org/10.1145/3079856.3080215
YOU?
·
· 2017
· Open Access
·
· DOI: https://doi.org/10.1145/3079856.3080215
As the size of Deep Neural Networks (DNNs) continues to grow to increase accuracy and solve more complex problems, their energy footprint also scales. Weight pruning reduces DNN model size and the computation by removing redundant weights. However, we implemented weight pruning for several popular networks on a variety of hardware platforms and observed surprising results. For many networks, the network sparsity caused by weight pruning will actually hurt the overall performance despite large reductions in the model size and required multiply-accumulate operations. Also, encoding the sparse format of pruned networks incurs additional storage space overhead. To overcome these challenges, we propose Scalpel that customizes DNN pruning to the underlying hardware by matching the pruned network structure to the data-parallel hardware organization. Scalpel consists of two techniques: SIMD-aware weight pruning and node pruning. For low-parallelism hardware (e.g., microcontroller), SIMD-aware weight pruning maintains weights in aligned fixed-size groups to fully utilize the SIMD units. For high-parallelism hardware (e.g., GPU), node pruning removes redundant nodes, not redundant weights, thereby reducing computation without sacrificing the dense matrix format. For hardware with moderate parallelism (e.g., desktop CPU), SIMD-aware weight pruning and node pruning are synergistically applied together. Across the microcontroller, CPU and GPU, Scalpel achieves mean speedups of 3.54x, 2.61x, and 1.25x while reducing the model sizes by 88%, 82%, and 53%. In comparison, traditional weight pruning achieves mean speedups of 1.90x, 1.06x, 0.41x across the three platforms.
Related Topics
Concepts
Metadata
- Type
- article
- Language
- en
- Landing Page
- https://doi.org/10.1145/3079856.3080215
- https://dl.acm.org/doi/pdf/10.1145/3079856.3080215?download=true
- OA Status
- gold
- Cited By
- 291
- References
- 50
- Related Works
- 10
- OpenAlex ID
- https://openalex.org/W2657126969
All OpenAlex metadata
Raw OpenAlex JSON
- OpenAlex ID
-
https://openalex.org/W2657126969Canonical identifier for this work in OpenAlex
- DOI
-
https://doi.org/10.1145/3079856.3080215Digital Object Identifier
- Title
-
ScalpelWork title
- Type
-
articleOpenAlex work type
- Language
-
enPrimary language
- Publication year
-
2017Year of publication
- Publication date
-
2017-06-24Full publication date if available
- Authors
-
Jiecao Yu, Andrew Lukefahr, David J. Palframan, Ganesh Dasika, Reetuparna Das, Scott MahlkeList of authors in order
- Landing page
-
https://doi.org/10.1145/3079856.3080215Publisher landing page
- PDF URL
-
https://dl.acm.org/doi/pdf/10.1145/3079856.3080215?download=trueDirect link to full text PDF
- Open access
-
YesWhether a free full text is available
- OA status
-
goldOpen access status per OpenAlex
- OA URL
-
https://dl.acm.org/doi/pdf/10.1145/3079856.3080215?download=trueDirect OA link when available
- Concepts
-
SIMD, Computer science, Pruning, Parallel computing, Computation, Node (physics), Overhead (engineering), Artificial neural network, Algorithm, Artificial intelligence, Engineering, Agronomy, Operating system, Biology, Structural engineeringTop concepts (fields/topics) attached by OpenAlex
- Cited by
-
291Total citation count in OpenAlex
- Citations by year (recent)
-
2025: 6, 2024: 6, 2023: 16, 2022: 9, 2021: 50Per-year citation counts (last 5 years)
- References (count)
-
50Number of works referenced by this work
- Related works (count)
-
10Other works algorithmically related by OpenAlex
Full payload
| id | https://openalex.org/W2657126969 |
|---|---|
| doi | https://doi.org/10.1145/3079856.3080215 |
| ids.doi | https://doi.org/10.1145/3079856.3080215 |
| ids.mag | 2657126969 |
| ids.openalex | https://openalex.org/W2657126969 |
| fwci | 25.93048514 |
| type | article |
| title | Scalpel |
| biblio.issue | |
| biblio.volume | |
| biblio.last_page | 560 |
| biblio.first_page | 548 |
| 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 | 0.9998999834060669 |
| 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/T10502 |
| topics[1].field.id | https://openalex.org/fields/22 |
| topics[1].field.display_name | Engineering |
| topics[1].score | 0.9997000098228455 |
| topics[1].domain.id | https://openalex.org/domains/3 |
| topics[1].domain.display_name | Physical Sciences |
| topics[1].subfield.id | https://openalex.org/subfields/2208 |
| topics[1].subfield.display_name | Electrical and Electronic Engineering |
| topics[1].display_name | Advanced Memory and Neural Computing |
| topics[2].id | https://openalex.org/T12808 |
| topics[2].field.id | https://openalex.org/fields/22 |
| topics[2].field.display_name | Engineering |
| topics[2].score | 0.9983999729156494 |
| topics[2].domain.id | https://openalex.org/domains/3 |
| topics[2].domain.display_name | Physical Sciences |
| topics[2].subfield.id | https://openalex.org/subfields/2208 |
| topics[2].subfield.display_name | Electrical and Electronic Engineering |
| topics[2].display_name | Ferroelectric and Negative Capacitance Devices |
| funders[0].id | https://openalex.org/F4320306076 |
| funders[0].ror | https://ror.org/021nxhr62 |
| funders[0].display_name | National Science Foundation |
| is_xpac | False |
| apc_list | |
| apc_paid | |
| concepts[0].id | https://openalex.org/C150552126 |
| concepts[0].level | 2 |
| concepts[0].score | 0.8298764824867249 |
| concepts[0].wikidata | https://www.wikidata.org/wiki/Q339387 |
| concepts[0].display_name | SIMD |
| concepts[1].id | https://openalex.org/C41008148 |
| concepts[1].level | 0 |
| concepts[1].score | 0.8256064653396606 |
| concepts[1].wikidata | https://www.wikidata.org/wiki/Q21198 |
| concepts[1].display_name | Computer science |
| concepts[2].id | https://openalex.org/C108010975 |
| concepts[2].level | 2 |
| concepts[2].score | 0.7658932209014893 |
| concepts[2].wikidata | https://www.wikidata.org/wiki/Q500094 |
| concepts[2].display_name | Pruning |
| concepts[3].id | https://openalex.org/C173608175 |
| concepts[3].level | 1 |
| concepts[3].score | 0.7430303692817688 |
| concepts[3].wikidata | https://www.wikidata.org/wiki/Q232661 |
| concepts[3].display_name | Parallel computing |
| concepts[4].id | https://openalex.org/C45374587 |
| concepts[4].level | 2 |
| concepts[4].score | 0.5039915442466736 |
| concepts[4].wikidata | https://www.wikidata.org/wiki/Q12525525 |
| concepts[4].display_name | Computation |
| concepts[5].id | https://openalex.org/C62611344 |
| concepts[5].level | 2 |
| concepts[5].score | 0.48880884051322937 |
| concepts[5].wikidata | https://www.wikidata.org/wiki/Q1062658 |
| concepts[5].display_name | Node (physics) |
| concepts[6].id | https://openalex.org/C2779960059 |
| concepts[6].level | 2 |
| concepts[6].score | 0.4440731406211853 |
| concepts[6].wikidata | https://www.wikidata.org/wiki/Q7113681 |
| concepts[6].display_name | Overhead (engineering) |
| concepts[7].id | https://openalex.org/C50644808 |
| concepts[7].level | 2 |
| concepts[7].score | 0.41165587306022644 |
| concepts[7].wikidata | https://www.wikidata.org/wiki/Q192776 |
| concepts[7].display_name | Artificial neural network |
| concepts[8].id | https://openalex.org/C11413529 |
| concepts[8].level | 1 |
| concepts[8].score | 0.33353865146636963 |
| concepts[8].wikidata | https://www.wikidata.org/wiki/Q8366 |
| concepts[8].display_name | Algorithm |
| concepts[9].id | https://openalex.org/C154945302 |
| concepts[9].level | 1 |
| concepts[9].score | 0.22331899404525757 |
| concepts[9].wikidata | https://www.wikidata.org/wiki/Q11660 |
| concepts[9].display_name | Artificial intelligence |
| concepts[10].id | https://openalex.org/C127413603 |
| concepts[10].level | 0 |
| concepts[10].score | 0.0 |
| concepts[10].wikidata | https://www.wikidata.org/wiki/Q11023 |
| concepts[10].display_name | Engineering |
| concepts[11].id | https://openalex.org/C6557445 |
| concepts[11].level | 1 |
| concepts[11].score | 0.0 |
| concepts[11].wikidata | https://www.wikidata.org/wiki/Q173113 |
| concepts[11].display_name | Agronomy |
| concepts[12].id | https://openalex.org/C111919701 |
| concepts[12].level | 1 |
| concepts[12].score | 0.0 |
| concepts[12].wikidata | https://www.wikidata.org/wiki/Q9135 |
| concepts[12].display_name | Operating system |
| concepts[13].id | https://openalex.org/C86803240 |
| concepts[13].level | 0 |
| concepts[13].score | 0.0 |
| concepts[13].wikidata | https://www.wikidata.org/wiki/Q420 |
| concepts[13].display_name | Biology |
| concepts[14].id | https://openalex.org/C66938386 |
| concepts[14].level | 1 |
| concepts[14].score | 0.0 |
| concepts[14].wikidata | https://www.wikidata.org/wiki/Q633538 |
| concepts[14].display_name | Structural engineering |
| keywords[0].id | https://openalex.org/keywords/simd |
| keywords[0].score | 0.8298764824867249 |
| keywords[0].display_name | SIMD |
| keywords[1].id | https://openalex.org/keywords/computer-science |
| keywords[1].score | 0.8256064653396606 |
| keywords[1].display_name | Computer science |
| keywords[2].id | https://openalex.org/keywords/pruning |
| keywords[2].score | 0.7658932209014893 |
| keywords[2].display_name | Pruning |
| keywords[3].id | https://openalex.org/keywords/parallel-computing |
| keywords[3].score | 0.7430303692817688 |
| keywords[3].display_name | Parallel computing |
| keywords[4].id | https://openalex.org/keywords/computation |
| keywords[4].score | 0.5039915442466736 |
| keywords[4].display_name | Computation |
| keywords[5].id | https://openalex.org/keywords/node |
| keywords[5].score | 0.48880884051322937 |
| keywords[5].display_name | Node (physics) |
| keywords[6].id | https://openalex.org/keywords/overhead |
| keywords[6].score | 0.4440731406211853 |
| keywords[6].display_name | Overhead (engineering) |
| keywords[7].id | https://openalex.org/keywords/artificial-neural-network |
| keywords[7].score | 0.41165587306022644 |
| keywords[7].display_name | Artificial neural network |
| keywords[8].id | https://openalex.org/keywords/algorithm |
| keywords[8].score | 0.33353865146636963 |
| keywords[8].display_name | Algorithm |
| keywords[9].id | https://openalex.org/keywords/artificial-intelligence |
| keywords[9].score | 0.22331899404525757 |
| keywords[9].display_name | Artificial intelligence |
| language | en |
| locations[0].id | doi:10.1145/3079856.3080215 |
| locations[0].is_oa | True |
| locations[0].source | |
| locations[0].license | |
| locations[0].pdf_url | https://dl.acm.org/doi/pdf/10.1145/3079856.3080215?download=true |
| locations[0].version | publishedVersion |
| locations[0].raw_type | proceedings-article |
| locations[0].license_id | |
| locations[0].is_accepted | True |
| locations[0].is_published | True |
| locations[0].raw_source_name | Proceedings of the 44th Annual International Symposium on Computer Architecture |
| locations[0].landing_page_url | https://doi.org/10.1145/3079856.3080215 |
| indexed_in | crossref |
| authorships[0].author.id | https://openalex.org/A5008163714 |
| authorships[0].author.orcid | https://orcid.org/0000-0003-2085-0312 |
| authorships[0].author.display_name | Jiecao Yu |
| authorships[0].countries | US |
| authorships[0].affiliations[0].institution_ids | https://openalex.org/I27837315 |
| authorships[0].affiliations[0].raw_affiliation_string | University of Michigan |
| authorships[0].institutions[0].id | https://openalex.org/I27837315 |
| authorships[0].institutions[0].ror | https://ror.org/00jmfr291 |
| authorships[0].institutions[0].type | education |
| authorships[0].institutions[0].lineage | https://openalex.org/I27837315 |
| authorships[0].institutions[0].country_code | US |
| authorships[0].institutions[0].display_name | University of Michigan |
| authorships[0].author_position | first |
| authorships[0].raw_author_name | Jiecao Yu |
| authorships[0].is_corresponding | False |
| authorships[0].raw_affiliation_strings | University of Michigan |
| authorships[1].author.id | https://openalex.org/A5070401389 |
| authorships[1].author.orcid | https://orcid.org/0009-0003-1167-3999 |
| authorships[1].author.display_name | Andrew Lukefahr |
| authorships[1].countries | US |
| authorships[1].affiliations[0].institution_ids | https://openalex.org/I27837315 |
| authorships[1].affiliations[0].raw_affiliation_string | University of Michigan |
| authorships[1].institutions[0].id | https://openalex.org/I27837315 |
| authorships[1].institutions[0].ror | https://ror.org/00jmfr291 |
| authorships[1].institutions[0].type | education |
| authorships[1].institutions[0].lineage | https://openalex.org/I27837315 |
| authorships[1].institutions[0].country_code | US |
| authorships[1].institutions[0].display_name | University of Michigan |
| authorships[1].author_position | middle |
| authorships[1].raw_author_name | Andrew Lukefahr |
| authorships[1].is_corresponding | False |
| authorships[1].raw_affiliation_strings | University of Michigan |
| authorships[2].author.id | https://openalex.org/A5044115982 |
| authorships[2].author.orcid | |
| authorships[2].author.display_name | David J. Palframan |
| authorships[2].countries | US |
| authorships[2].affiliations[0].institution_ids | https://openalex.org/I4210156213 |
| authorships[2].affiliations[0].raw_affiliation_string | ARM |
| authorships[2].institutions[0].id | https://openalex.org/I4210156213 |
| authorships[2].institutions[0].ror | https://ror.org/05vfrxy92 |
| authorships[2].institutions[0].type | nonprofit |
| authorships[2].institutions[0].lineage | https://openalex.org/I4210156213 |
| authorships[2].institutions[0].country_code | US |
| authorships[2].institutions[0].display_name | American Rock Mechanics Association |
| authorships[2].author_position | middle |
| authorships[2].raw_author_name | David Palframan |
| authorships[2].is_corresponding | False |
| authorships[2].raw_affiliation_strings | ARM |
| authorships[3].author.id | https://openalex.org/A5004292742 |
| authorships[3].author.orcid | |
| authorships[3].author.display_name | Ganesh Dasika |
| authorships[3].countries | US |
| authorships[3].affiliations[0].institution_ids | https://openalex.org/I4210156213 |
| authorships[3].affiliations[0].raw_affiliation_string | ARM |
| authorships[3].institutions[0].id | https://openalex.org/I4210156213 |
| authorships[3].institutions[0].ror | https://ror.org/05vfrxy92 |
| authorships[3].institutions[0].type | nonprofit |
| authorships[3].institutions[0].lineage | https://openalex.org/I4210156213 |
| authorships[3].institutions[0].country_code | US |
| authorships[3].institutions[0].display_name | American Rock Mechanics Association |
| authorships[3].author_position | middle |
| authorships[3].raw_author_name | Ganesh Dasika |
| authorships[3].is_corresponding | False |
| authorships[3].raw_affiliation_strings | ARM |
| authorships[4].author.id | https://openalex.org/A5027544576 |
| authorships[4].author.orcid | https://orcid.org/0000-0002-5894-8342 |
| authorships[4].author.display_name | Reetuparna Das |
| authorships[4].countries | US |
| authorships[4].affiliations[0].institution_ids | https://openalex.org/I27837315 |
| authorships[4].affiliations[0].raw_affiliation_string | University of Michigan |
| authorships[4].institutions[0].id | https://openalex.org/I27837315 |
| authorships[4].institutions[0].ror | https://ror.org/00jmfr291 |
| authorships[4].institutions[0].type | education |
| authorships[4].institutions[0].lineage | https://openalex.org/I27837315 |
| authorships[4].institutions[0].country_code | US |
| authorships[4].institutions[0].display_name | University of Michigan |
| authorships[4].author_position | middle |
| authorships[4].raw_author_name | Reetuparna Das |
| authorships[4].is_corresponding | False |
| authorships[4].raw_affiliation_strings | University of Michigan |
| authorships[5].author.id | https://openalex.org/A5002075773 |
| authorships[5].author.orcid | https://orcid.org/0000-0002-0438-0616 |
| authorships[5].author.display_name | Scott Mahlke |
| authorships[5].countries | US |
| authorships[5].affiliations[0].institution_ids | https://openalex.org/I27837315 |
| authorships[5].affiliations[0].raw_affiliation_string | University of Michigan |
| authorships[5].institutions[0].id | https://openalex.org/I27837315 |
| authorships[5].institutions[0].ror | https://ror.org/00jmfr291 |
| authorships[5].institutions[0].type | education |
| authorships[5].institutions[0].lineage | https://openalex.org/I27837315 |
| authorships[5].institutions[0].country_code | US |
| authorships[5].institutions[0].display_name | University of Michigan |
| authorships[5].author_position | last |
| authorships[5].raw_author_name | Scott Mahlke |
| authorships[5].is_corresponding | False |
| authorships[5].raw_affiliation_strings | University of Michigan |
| has_content.pdf | True |
| has_content.grobid_xml | True |
| is_paratext | False |
| open_access.is_oa | True |
| open_access.oa_url | https://dl.acm.org/doi/pdf/10.1145/3079856.3080215?download=true |
| open_access.oa_status | gold |
| open_access.any_repository_has_fulltext | False |
| created_date | 2017-06-30T00:00:00 |
| display_name | Scalpel |
| has_fulltext | False |
| is_retracted | False |
| updated_date | 2025-11-23T05:10:03.516525 |
| 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 | 0.9998999834060669 |
| 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/W2291920536, https://openalex.org/W2022397046, https://openalex.org/W2162726111, https://openalex.org/W1603583590, https://openalex.org/W338671845, https://openalex.org/W1587248296, https://openalex.org/W2117755361, https://openalex.org/W2286348849, https://openalex.org/W2063770303, https://openalex.org/W2018511057 |
| cited_by_count | 291 |
| counts_by_year[0].year | 2025 |
| counts_by_year[0].cited_by_count | 6 |
| counts_by_year[1].year | 2024 |
| counts_by_year[1].cited_by_count | 6 |
| counts_by_year[2].year | 2023 |
| counts_by_year[2].cited_by_count | 16 |
| counts_by_year[3].year | 2022 |
| counts_by_year[3].cited_by_count | 9 |
| counts_by_year[4].year | 2021 |
| counts_by_year[4].cited_by_count | 50 |
| counts_by_year[5].year | 2020 |
| counts_by_year[5].cited_by_count | 70 |
| counts_by_year[6].year | 2019 |
| counts_by_year[6].cited_by_count | 77 |
| counts_by_year[7].year | 2018 |
| counts_by_year[7].cited_by_count | 51 |
| counts_by_year[8].year | 2017 |
| counts_by_year[8].cited_by_count | 6 |
| locations_count | 1 |
| best_oa_location.id | doi:10.1145/3079856.3080215 |
| best_oa_location.is_oa | True |
| best_oa_location.source | |
| best_oa_location.license | |
| best_oa_location.pdf_url | https://dl.acm.org/doi/pdf/10.1145/3079856.3080215?download=true |
| best_oa_location.version | publishedVersion |
| best_oa_location.raw_type | proceedings-article |
| best_oa_location.license_id | |
| best_oa_location.is_accepted | True |
| best_oa_location.is_published | True |
| best_oa_location.raw_source_name | Proceedings of the 44th Annual International Symposium on Computer Architecture |
| best_oa_location.landing_page_url | https://doi.org/10.1145/3079856.3080215 |
| primary_location.id | doi:10.1145/3079856.3080215 |
| primary_location.is_oa | True |
| primary_location.source | |
| primary_location.license | |
| primary_location.pdf_url | https://dl.acm.org/doi/pdf/10.1145/3079856.3080215?download=true |
| primary_location.version | publishedVersion |
| primary_location.raw_type | proceedings-article |
| primary_location.license_id | |
| primary_location.is_accepted | True |
| primary_location.is_published | True |
| primary_location.raw_source_name | Proceedings of the 44th Annual International Symposium on Computer Architecture |
| primary_location.landing_page_url | https://doi.org/10.1145/3079856.3080215 |
| publication_date | 2017-06-24 |
| publication_year | 2017 |
| referenced_works | https://openalex.org/W4251575795, https://openalex.org/W2034940213, https://openalex.org/W4212788319, https://openalex.org/W2048266589, https://openalex.org/W2442974303, https://openalex.org/W4254672563, https://openalex.org/W2117130368, https://openalex.org/W2067523571, https://openalex.org/W1536680647, https://openalex.org/W2950656546, https://openalex.org/W2156150815, https://openalex.org/W2084910356, https://openalex.org/W2155893237, https://openalex.org/W2112796928, https://openalex.org/W4251607986, https://openalex.org/W4245199738, https://openalex.org/W4243519499, https://openalex.org/W2097117768, https://openalex.org/W2010291496, https://openalex.org/W2963981420, https://openalex.org/W2194775991, https://openalex.org/W2618530766, https://openalex.org/W2508602506, https://openalex.org/W587794757, https://openalex.org/W1789336918, https://openalex.org/W1841592590, https://openalex.org/W2621285457, https://openalex.org/W2518281301, https://openalex.org/W2114766824, https://openalex.org/W2463256602, https://openalex.org/W2516141709, https://openalex.org/W1686810756, https://openalex.org/W2152839228, https://openalex.org/W1821462560, https://openalex.org/W2556370608, https://openalex.org/W1667652561, https://openalex.org/W2511743527, https://openalex.org/W2095705004, https://openalex.org/W2513554817, https://openalex.org/W2893550440, https://openalex.org/W2483966489, https://openalex.org/W2504108613, https://openalex.org/W2186615578, https://openalex.org/W2260663238, https://openalex.org/W2161591461, https://openalex.org/W2952881492, https://openalex.org/W2134797427, https://openalex.org/W1799366690, https://openalex.org/W2172654076, https://openalex.org/W2963674932 |
| referenced_works_count | 50 |
| abstract_inverted_index.a | 47 |
| abstract_inverted_index.As | 0 |
| abstract_inverted_index.In | 218 |
| abstract_inverted_index.To | 96 |
| abstract_inverted_index.by | 33, 63, 111, 213 |
| abstract_inverted_index.in | 75, 143 |
| abstract_inverted_index.of | 3, 49, 88, 124, 203, 226 |
| abstract_inverted_index.on | 46 |
| abstract_inverted_index.to | 9, 11, 107, 117, 147 |
| abstract_inverted_index.we | 38, 100 |
| abstract_inverted_index.CPU | 196 |
| abstract_inverted_index.DNN | 27, 105 |
| abstract_inverted_index.For | 56, 133, 153, 175 |
| abstract_inverted_index.and | 14, 30, 52, 79, 130, 186, 197, 206, 216 |
| abstract_inverted_index.are | 189 |
| abstract_inverted_index.for | 42 |
| abstract_inverted_index.not | 163 |
| abstract_inverted_index.the | 1, 31, 59, 69, 76, 85, 108, 113, 118, 150, 171, 194, 210, 231 |
| abstract_inverted_index.two | 125 |
| abstract_inverted_index.53%. | 217 |
| abstract_inverted_index.82%, | 215 |
| abstract_inverted_index.88%, | 214 |
| abstract_inverted_index.Deep | 4 |
| abstract_inverted_index.GPU, | 198 |
| abstract_inverted_index.SIMD | 151 |
| abstract_inverted_index.also | 22 |
| abstract_inverted_index.grow | 10 |
| abstract_inverted_index.hurt | 68 |
| abstract_inverted_index.many | 57 |
| abstract_inverted_index.mean | 201, 224 |
| abstract_inverted_index.more | 16 |
| abstract_inverted_index.node | 131, 158, 187 |
| abstract_inverted_index.size | 2, 29, 78 |
| abstract_inverted_index.that | 103 |
| abstract_inverted_index.will | 66 |
| abstract_inverted_index.with | 177 |
| abstract_inverted_index.0.41x | 229 |
| abstract_inverted_index.1.25x | 207 |
| abstract_inverted_index.Also, | 83 |
| abstract_inverted_index.CPU), | 182 |
| abstract_inverted_index.GPU), | 157 |
| abstract_inverted_index.dense | 172 |
| abstract_inverted_index.fully | 148 |
| abstract_inverted_index.large | 73 |
| abstract_inverted_index.model | 28, 77, 211 |
| abstract_inverted_index.sizes | 212 |
| abstract_inverted_index.solve | 15 |
| abstract_inverted_index.space | 94 |
| abstract_inverted_index.their | 19 |
| abstract_inverted_index.these | 98 |
| abstract_inverted_index.three | 232 |
| abstract_inverted_index.while | 208 |
| abstract_inverted_index.(DNNs) | 7 |
| abstract_inverted_index.(e.g., | 136, 156, 180 |
| abstract_inverted_index.1.06x, | 228 |
| abstract_inverted_index.1.90x, | 227 |
| abstract_inverted_index.2.61x, | 205 |
| abstract_inverted_index.3.54x, | 204 |
| abstract_inverted_index.Across | 193 |
| abstract_inverted_index.Neural | 5 |
| abstract_inverted_index.Weight | 24 |
| abstract_inverted_index.across | 230 |
| abstract_inverted_index.caused | 62 |
| abstract_inverted_index.energy | 20 |
| abstract_inverted_index.format | 87 |
| abstract_inverted_index.groups | 146 |
| abstract_inverted_index.incurs | 91 |
| abstract_inverted_index.matrix | 173 |
| abstract_inverted_index.nodes, | 162 |
| abstract_inverted_index.pruned | 89, 114 |
| abstract_inverted_index.sparse | 86 |
| abstract_inverted_index.units. | 152 |
| abstract_inverted_index.weight | 40, 64, 128, 139, 184, 221 |
| abstract_inverted_index.Scalpel | 102, 122, 199 |
| abstract_inverted_index.aligned | 144 |
| abstract_inverted_index.applied | 191 |
| abstract_inverted_index.complex | 17 |
| abstract_inverted_index.desktop | 181 |
| abstract_inverted_index.despite | 72 |
| abstract_inverted_index.format. | 174 |
| abstract_inverted_index.network | 60, 115 |
| abstract_inverted_index.overall | 70 |
| abstract_inverted_index.popular | 44 |
| abstract_inverted_index.propose | 101 |
| abstract_inverted_index.pruning | 25, 41, 65, 106, 129, 140, 159, 185, 188, 222 |
| abstract_inverted_index.reduces | 26 |
| abstract_inverted_index.removes | 160 |
| abstract_inverted_index.scales. | 23 |
| abstract_inverted_index.several | 43 |
| abstract_inverted_index.storage | 93 |
| abstract_inverted_index.thereby | 166 |
| abstract_inverted_index.utilize | 149 |
| abstract_inverted_index.variety | 48 |
| abstract_inverted_index.weights | 142 |
| abstract_inverted_index.without | 169 |
| abstract_inverted_index.However, | 37 |
| abstract_inverted_index.Networks | 6 |
| abstract_inverted_index.accuracy | 13 |
| abstract_inverted_index.achieves | 200, 223 |
| abstract_inverted_index.actually | 67 |
| abstract_inverted_index.consists | 123 |
| abstract_inverted_index.encoding | 84 |
| abstract_inverted_index.hardware | 50, 110, 120, 135, 155, 176 |
| abstract_inverted_index.increase | 12 |
| abstract_inverted_index.matching | 112 |
| abstract_inverted_index.moderate | 178 |
| abstract_inverted_index.networks | 45, 90 |
| abstract_inverted_index.observed | 53 |
| abstract_inverted_index.overcome | 97 |
| abstract_inverted_index.pruning. | 132 |
| abstract_inverted_index.reducing | 167, 209 |
| abstract_inverted_index.removing | 34 |
| abstract_inverted_index.required | 80 |
| abstract_inverted_index.results. | 55 |
| abstract_inverted_index.sparsity | 61 |
| abstract_inverted_index.speedups | 202, 225 |
| abstract_inverted_index.weights, | 165 |
| abstract_inverted_index.weights. | 36 |
| abstract_inverted_index.continues | 8 |
| abstract_inverted_index.footprint | 21 |
| abstract_inverted_index.maintains | 141 |
| abstract_inverted_index.networks, | 58 |
| abstract_inverted_index.overhead. | 95 |
| abstract_inverted_index.platforms | 51 |
| abstract_inverted_index.problems, | 18 |
| abstract_inverted_index.redundant | 35, 161, 164 |
| abstract_inverted_index.structure | 116 |
| abstract_inverted_index.together. | 192 |
| abstract_inverted_index.SIMD-aware | 127, 138, 183 |
| abstract_inverted_index.additional | 92 |
| abstract_inverted_index.customizes | 104 |
| abstract_inverted_index.fixed-size | 145 |
| abstract_inverted_index.platforms. | 233 |
| abstract_inverted_index.reductions | 74 |
| abstract_inverted_index.surprising | 54 |
| abstract_inverted_index.underlying | 109 |
| abstract_inverted_index.challenges, | 99 |
| abstract_inverted_index.comparison, | 219 |
| abstract_inverted_index.computation | 32, 168 |
| abstract_inverted_index.implemented | 39 |
| abstract_inverted_index.operations. | 82 |
| abstract_inverted_index.parallelism | 179 |
| abstract_inverted_index.performance | 71 |
| abstract_inverted_index.sacrificing | 170 |
| abstract_inverted_index.techniques: | 126 |
| abstract_inverted_index.traditional | 220 |
| abstract_inverted_index.data-parallel | 119 |
| abstract_inverted_index.organization. | 121 |
| abstract_inverted_index.low-parallelism | 134 |
| abstract_inverted_index.synergistically | 190 |
| abstract_inverted_index.high-parallelism | 154 |
| abstract_inverted_index.microcontroller, | 195 |
| abstract_inverted_index.microcontroller), | 137 |
| abstract_inverted_index.multiply-accumulate | 81 |
| cited_by_percentile_year.max | 100 |
| cited_by_percentile_year.min | 98 |
| countries_distinct_count | 1 |
| institutions_distinct_count | 6 |
| sustainable_development_goals[0].id | https://metadata.un.org/sdg/7 |
| sustainable_development_goals[0].score | 0.8999999761581421 |
| sustainable_development_goals[0].display_name | Affordable and clean energy |
| citation_normalized_percentile.value | 0.99444876 |
| citation_normalized_percentile.is_in_top_1_percent | True |
| citation_normalized_percentile.is_in_top_10_percent | True |