Bit-serial Weight Pools: Compression and Arbitrary Precision Execution of Neural Networks on Resource Constrained Processors Article Swipe
Shurui Li
,
Puneet Gupta
·
YOU?
·
· 2022
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2201.11651
YOU?
·
· 2022
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2201.11651
Applications of neural networks on edge systems have proliferated in recent years but the ever-increasing model size makes neural networks not able to deploy on resource-constrained microcontrollers efficiently. We propose bit-serial weight pools, an end-to-end framework that includes network compression and acceleration of arbitrary sub-byte precision. The framework can achieve up to 8x compression compared to 8-bit networks by sharing a pool of weights across the entire network. We further propose a bit-serial lookup based software implementation that allows runtime-bitwidth tradeoff and is able to achieve more than 2.8x speedup and 7.5x storage compression compared to 8-bit weight pool networks, with less than 1% accuracy drop.
Related Topics
Concepts
Metadata
- Type
- preprint
- Language
- en
- Landing Page
- http://arxiv.org/abs/2201.11651
- https://arxiv.org/pdf/2201.11651
- OA Status
- green
- Cited By
- 2
- Related Works
- 10
- OpenAlex ID
- https://openalex.org/W4221142407
All OpenAlex metadata
Raw OpenAlex JSON
- OpenAlex ID
-
https://openalex.org/W4221142407Canonical identifier for this work in OpenAlex
- DOI
-
https://doi.org/10.48550/arxiv.2201.11651Digital Object Identifier
- Title
-
Bit-serial Weight Pools: Compression and Arbitrary Precision Execution of Neural Networks on Resource Constrained ProcessorsWork title
- Type
-
preprintOpenAlex work type
- Language
-
enPrimary language
- Publication year
-
2022Year of publication
- Publication date
-
2022-01-25Full publication date if available
- Authors
-
Shurui Li, Puneet GuptaList of authors in order
- Landing page
-
https://arxiv.org/abs/2201.11651Publisher landing page
- PDF URL
-
https://arxiv.org/pdf/2201.11651Direct 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/2201.11651Direct OA link when available
- Concepts
-
Computer science, Speedup, Artificial neural network, Byte, Parallel computing, Compression (physics), Data compression, Enhanced Data Rates for GSM Evolution, Lookup table, Algorithm, Computer hardware, Operating system, Artificial intelligence, Materials science, Composite materialTop concepts (fields/topics) attached by OpenAlex
- Cited by
-
2Total citation count in OpenAlex
- Citations by year (recent)
-
2023: 2Per-year citation counts (last 5 years)
- Related works (count)
-
10Other works algorithmically related by OpenAlex
Full payload
| id | https://openalex.org/W4221142407 |
|---|---|
| doi | https://doi.org/10.48550/arxiv.2201.11651 |
| ids.doi | https://doi.org/10.48550/arxiv.2201.11651 |
| ids.openalex | https://openalex.org/W4221142407 |
| fwci | |
| type | preprint |
| title | Bit-serial Weight Pools: Compression and Arbitrary Precision Execution of Neural Networks on Resource Constrained Processors |
| biblio.issue | |
| biblio.volume | |
| biblio.last_page | |
| biblio.first_page | |
| topics[0].id | https://openalex.org/T10054 |
| topics[0].field.id | https://openalex.org/fields/17 |
| topics[0].field.display_name | Computer Science |
| topics[0].score | 0.9973000288009644 |
| topics[0].domain.id | https://openalex.org/domains/3 |
| topics[0].domain.display_name | Physical Sciences |
| topics[0].subfield.id | https://openalex.org/subfields/1708 |
| topics[0].subfield.display_name | Hardware and Architecture |
| topics[0].display_name | Parallel Computing and Optimization Techniques |
| topics[1].id | https://openalex.org/T10320 |
| topics[1].field.id | https://openalex.org/fields/17 |
| topics[1].field.display_name | Computer Science |
| topics[1].score | 0.9879999756813049 |
| topics[1].domain.id | https://openalex.org/domains/3 |
| topics[1].domain.display_name | Physical Sciences |
| topics[1].subfield.id | https://openalex.org/subfields/1702 |
| topics[1].subfield.display_name | Artificial Intelligence |
| topics[1].display_name | Neural Networks and Applications |
| topics[2].id | https://openalex.org/T11697 |
| topics[2].field.id | https://openalex.org/fields/17 |
| topics[2].field.display_name | Computer Science |
| topics[2].score | 0.9807999730110168 |
| topics[2].domain.id | https://openalex.org/domains/3 |
| topics[2].domain.display_name | Physical Sciences |
| topics[2].subfield.id | https://openalex.org/subfields/1703 |
| topics[2].subfield.display_name | Computational Theory and Mathematics |
| topics[2].display_name | Numerical Methods and Algorithms |
| is_xpac | False |
| apc_list | |
| apc_paid | |
| concepts[0].id | https://openalex.org/C41008148 |
| concepts[0].level | 0 |
| concepts[0].score | 0.833977222442627 |
| concepts[0].wikidata | https://www.wikidata.org/wiki/Q21198 |
| concepts[0].display_name | Computer science |
| concepts[1].id | https://openalex.org/C68339613 |
| concepts[1].level | 2 |
| concepts[1].score | 0.5929080843925476 |
| concepts[1].wikidata | https://www.wikidata.org/wiki/Q1549489 |
| concepts[1].display_name | Speedup |
| concepts[2].id | https://openalex.org/C50644808 |
| concepts[2].level | 2 |
| concepts[2].score | 0.5603628158569336 |
| concepts[2].wikidata | https://www.wikidata.org/wiki/Q192776 |
| concepts[2].display_name | Artificial neural network |
| concepts[3].id | https://openalex.org/C43364308 |
| concepts[3].level | 2 |
| concepts[3].score | 0.5472280383110046 |
| concepts[3].wikidata | https://www.wikidata.org/wiki/Q8799 |
| concepts[3].display_name | Byte |
| concepts[4].id | https://openalex.org/C173608175 |
| concepts[4].level | 1 |
| concepts[4].score | 0.5383990406990051 |
| concepts[4].wikidata | https://www.wikidata.org/wiki/Q232661 |
| concepts[4].display_name | Parallel computing |
| concepts[5].id | https://openalex.org/C180016635 |
| concepts[5].level | 2 |
| concepts[5].score | 0.4947311580181122 |
| concepts[5].wikidata | https://www.wikidata.org/wiki/Q2712821 |
| concepts[5].display_name | Compression (physics) |
| concepts[6].id | https://openalex.org/C78548338 |
| concepts[6].level | 2 |
| concepts[6].score | 0.4555444121360779 |
| concepts[6].wikidata | https://www.wikidata.org/wiki/Q2493 |
| concepts[6].display_name | Data compression |
| concepts[7].id | https://openalex.org/C162307627 |
| concepts[7].level | 2 |
| concepts[7].score | 0.43855342268943787 |
| concepts[7].wikidata | https://www.wikidata.org/wiki/Q204833 |
| concepts[7].display_name | Enhanced Data Rates for GSM Evolution |
| concepts[8].id | https://openalex.org/C134835016 |
| concepts[8].level | 2 |
| concepts[8].score | 0.41558128595352173 |
| concepts[8].wikidata | https://www.wikidata.org/wiki/Q690265 |
| concepts[8].display_name | Lookup table |
| concepts[9].id | https://openalex.org/C11413529 |
| concepts[9].level | 1 |
| concepts[9].score | 0.39190196990966797 |
| concepts[9].wikidata | https://www.wikidata.org/wiki/Q8366 |
| concepts[9].display_name | Algorithm |
| concepts[10].id | https://openalex.org/C9390403 |
| concepts[10].level | 1 |
| concepts[10].score | 0.2781105041503906 |
| concepts[10].wikidata | https://www.wikidata.org/wiki/Q3966 |
| concepts[10].display_name | Computer hardware |
| concepts[11].id | https://openalex.org/C111919701 |
| concepts[11].level | 1 |
| concepts[11].score | 0.1511901617050171 |
| concepts[11].wikidata | https://www.wikidata.org/wiki/Q9135 |
| concepts[11].display_name | Operating system |
| concepts[12].id | https://openalex.org/C154945302 |
| concepts[12].level | 1 |
| concepts[12].score | 0.08607557415962219 |
| concepts[12].wikidata | https://www.wikidata.org/wiki/Q11660 |
| concepts[12].display_name | Artificial intelligence |
| concepts[13].id | https://openalex.org/C192562407 |
| concepts[13].level | 0 |
| concepts[13].score | 0.0 |
| concepts[13].wikidata | https://www.wikidata.org/wiki/Q228736 |
| concepts[13].display_name | Materials science |
| concepts[14].id | https://openalex.org/C159985019 |
| concepts[14].level | 1 |
| concepts[14].score | 0.0 |
| concepts[14].wikidata | https://www.wikidata.org/wiki/Q181790 |
| concepts[14].display_name | Composite material |
| keywords[0].id | https://openalex.org/keywords/computer-science |
| keywords[0].score | 0.833977222442627 |
| keywords[0].display_name | Computer science |
| keywords[1].id | https://openalex.org/keywords/speedup |
| keywords[1].score | 0.5929080843925476 |
| keywords[1].display_name | Speedup |
| keywords[2].id | https://openalex.org/keywords/artificial-neural-network |
| keywords[2].score | 0.5603628158569336 |
| keywords[2].display_name | Artificial neural network |
| keywords[3].id | https://openalex.org/keywords/byte |
| keywords[3].score | 0.5472280383110046 |
| keywords[3].display_name | Byte |
| keywords[4].id | https://openalex.org/keywords/parallel-computing |
| keywords[4].score | 0.5383990406990051 |
| keywords[4].display_name | Parallel computing |
| keywords[5].id | https://openalex.org/keywords/compression |
| keywords[5].score | 0.4947311580181122 |
| keywords[5].display_name | Compression (physics) |
| keywords[6].id | https://openalex.org/keywords/data-compression |
| keywords[6].score | 0.4555444121360779 |
| keywords[6].display_name | Data compression |
| keywords[7].id | https://openalex.org/keywords/enhanced-data-rates-for-gsm-evolution |
| keywords[7].score | 0.43855342268943787 |
| keywords[7].display_name | Enhanced Data Rates for GSM Evolution |
| keywords[8].id | https://openalex.org/keywords/lookup-table |
| keywords[8].score | 0.41558128595352173 |
| keywords[8].display_name | Lookup table |
| keywords[9].id | https://openalex.org/keywords/algorithm |
| keywords[9].score | 0.39190196990966797 |
| keywords[9].display_name | Algorithm |
| keywords[10].id | https://openalex.org/keywords/computer-hardware |
| keywords[10].score | 0.2781105041503906 |
| keywords[10].display_name | Computer hardware |
| keywords[11].id | https://openalex.org/keywords/operating-system |
| keywords[11].score | 0.1511901617050171 |
| keywords[11].display_name | Operating system |
| keywords[12].id | https://openalex.org/keywords/artificial-intelligence |
| keywords[12].score | 0.08607557415962219 |
| keywords[12].display_name | Artificial intelligence |
| language | en |
| locations[0].id | pmh:oai:arXiv.org:2201.11651 |
| 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/2201.11651 |
| locations[0].version | submittedVersion |
| locations[0].raw_type | |
| 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/2201.11651 |
| locations[1].id | doi:10.48550/arxiv.2201.11651 |
| 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 | cc-by |
| locations[1].pdf_url | |
| locations[1].version | |
| locations[1].raw_type | article |
| locations[1].license_id | https://openalex.org/licenses/cc-by |
| locations[1].is_accepted | False |
| locations[1].is_published | |
| locations[1].raw_source_name | |
| locations[1].landing_page_url | https://doi.org/10.48550/arxiv.2201.11651 |
| indexed_in | arxiv, datacite |
| authorships[0].author.id | https://openalex.org/A5012719636 |
| authorships[0].author.orcid | https://orcid.org/0000-0001-6441-3988 |
| authorships[0].author.display_name | Shurui Li |
| authorships[0].author_position | first |
| authorships[0].raw_author_name | Li, Shurui |
| authorships[0].is_corresponding | False |
| authorships[1].author.id | https://openalex.org/A5084229134 |
| authorships[1].author.orcid | https://orcid.org/0000-0002-6188-1134 |
| authorships[1].author.display_name | Puneet Gupta |
| authorships[1].author_position | last |
| authorships[1].raw_author_name | Gupta, Puneet |
| authorships[1].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/2201.11651 |
| open_access.oa_status | green |
| open_access.any_repository_has_fulltext | False |
| created_date | 2025-10-10T00:00:00 |
| display_name | Bit-serial Weight Pools: Compression and Arbitrary Precision Execution of Neural Networks on Resource Constrained Processors |
| has_fulltext | False |
| is_retracted | False |
| updated_date | 2025-11-06T06:51:31.235846 |
| primary_topic.id | https://openalex.org/T10054 |
| primary_topic.field.id | https://openalex.org/fields/17 |
| primary_topic.field.display_name | Computer Science |
| primary_topic.score | 0.9973000288009644 |
| primary_topic.domain.id | https://openalex.org/domains/3 |
| primary_topic.domain.display_name | Physical Sciences |
| primary_topic.subfield.id | https://openalex.org/subfields/1708 |
| primary_topic.subfield.display_name | Hardware and Architecture |
| primary_topic.display_name | Parallel Computing and Optimization Techniques |
| related_works | https://openalex.org/W2058965144, https://openalex.org/W2164382479, https://openalex.org/W2146343568, https://openalex.org/W98480971, https://openalex.org/W2150291671, https://openalex.org/W2013643406, https://openalex.org/W3044766408, https://openalex.org/W2138567775, https://openalex.org/W2612632602, https://openalex.org/W2321805087 |
| cited_by_count | 2 |
| counts_by_year[0].year | 2023 |
| counts_by_year[0].cited_by_count | 2 |
| locations_count | 2 |
| best_oa_location.id | pmh:oai:arXiv.org:2201.11651 |
| 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/2201.11651 |
| best_oa_location.version | submittedVersion |
| best_oa_location.raw_type | |
| 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/2201.11651 |
| primary_location.id | pmh:oai:arXiv.org:2201.11651 |
| 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/2201.11651 |
| primary_location.version | submittedVersion |
| primary_location.raw_type | |
| 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/2201.11651 |
| publication_date | 2022-01-25 |
| publication_year | 2022 |
| referenced_works_count | 0 |
| abstract_inverted_index.a | 60, 71 |
| abstract_inverted_index.1% | 103 |
| abstract_inverted_index.8x | 52 |
| abstract_inverted_index.We | 28, 68 |
| abstract_inverted_index.an | 33 |
| abstract_inverted_index.by | 58 |
| abstract_inverted_index.in | 9 |
| abstract_inverted_index.is | 82 |
| abstract_inverted_index.of | 1, 42, 62 |
| abstract_inverted_index.on | 4, 24 |
| abstract_inverted_index.to | 22, 51, 55, 84, 95 |
| abstract_inverted_index.up | 50 |
| abstract_inverted_index.The | 46 |
| abstract_inverted_index.and | 40, 81, 90 |
| abstract_inverted_index.but | 12 |
| abstract_inverted_index.can | 48 |
| abstract_inverted_index.not | 20 |
| abstract_inverted_index.the | 13, 65 |
| abstract_inverted_index.2.8x | 88 |
| abstract_inverted_index.7.5x | 91 |
| abstract_inverted_index.able | 21, 83 |
| abstract_inverted_index.edge | 5 |
| abstract_inverted_index.have | 7 |
| abstract_inverted_index.less | 101 |
| abstract_inverted_index.more | 86 |
| abstract_inverted_index.pool | 61, 98 |
| abstract_inverted_index.size | 16 |
| abstract_inverted_index.than | 87, 102 |
| abstract_inverted_index.that | 36, 77 |
| abstract_inverted_index.with | 100 |
| abstract_inverted_index.8-bit | 56, 96 |
| abstract_inverted_index.based | 74 |
| abstract_inverted_index.drop. | 105 |
| abstract_inverted_index.makes | 17 |
| abstract_inverted_index.model | 15 |
| abstract_inverted_index.years | 11 |
| abstract_inverted_index.across | 64 |
| abstract_inverted_index.allows | 78 |
| abstract_inverted_index.deploy | 23 |
| abstract_inverted_index.entire | 66 |
| abstract_inverted_index.lookup | 73 |
| abstract_inverted_index.neural | 2, 18 |
| abstract_inverted_index.pools, | 32 |
| abstract_inverted_index.recent | 10 |
| abstract_inverted_index.weight | 31, 97 |
| abstract_inverted_index.achieve | 49, 85 |
| abstract_inverted_index.further | 69 |
| abstract_inverted_index.network | 38 |
| abstract_inverted_index.propose | 29, 70 |
| abstract_inverted_index.sharing | 59 |
| abstract_inverted_index.speedup | 89 |
| abstract_inverted_index.storage | 92 |
| abstract_inverted_index.systems | 6 |
| abstract_inverted_index.weights | 63 |
| abstract_inverted_index.accuracy | 104 |
| abstract_inverted_index.compared | 54, 94 |
| abstract_inverted_index.includes | 37 |
| abstract_inverted_index.network. | 67 |
| abstract_inverted_index.networks | 3, 19, 57 |
| abstract_inverted_index.software | 75 |
| abstract_inverted_index.sub-byte | 44 |
| abstract_inverted_index.tradeoff | 80 |
| abstract_inverted_index.arbitrary | 43 |
| abstract_inverted_index.framework | 35, 47 |
| abstract_inverted_index.networks, | 99 |
| abstract_inverted_index.bit-serial | 30, 72 |
| abstract_inverted_index.end-to-end | 34 |
| abstract_inverted_index.precision. | 45 |
| abstract_inverted_index.compression | 39, 53, 93 |
| abstract_inverted_index.Applications | 0 |
| abstract_inverted_index.acceleration | 41 |
| abstract_inverted_index.efficiently. | 27 |
| abstract_inverted_index.proliferated | 8 |
| abstract_inverted_index.implementation | 76 |
| abstract_inverted_index.ever-increasing | 14 |
| abstract_inverted_index.microcontrollers | 26 |
| abstract_inverted_index.runtime-bitwidth | 79 |
| abstract_inverted_index.resource-constrained | 25 |
| cited_by_percentile_year | |
| countries_distinct_count | 0 |
| institutions_distinct_count | 2 |
| citation_normalized_percentile |