Umeå University's logo

umu.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Density Encoding Enables Resource-Efficient Randomly Connected Neural Networks
Redwood Center for Theoretical Neuroscience, University of California at Berkeley, CA, Berkeley, United States; Intelligent Systems Lab, Research Institutes of Sweden, Kista, Sweden.
Netlight Consulting AB, Stockholm, Sweden.
Redwood Center for Theoretical Neuroscience, University of California at Berkeley, CA, Berkeley, United States.
Umeå University, Faculty of Medicine, Department of Radiation Sciences, Radiation Physics.ORCID iD: 0000-0002-1313-0934
Show others and affiliations
2021 (English)In: IEEE Transactions on Neural Networks and Learning Systems, ISSN 2162-237X, E-ISSN 2162-2388, Vol. 32, no 8, p. 3777-3783, article id 9174774Article in journal (Refereed) Published
Abstract [en]

The deployment of machine learning algorithms on resource-constrained edge devices is an important challenge from both theoretical and applied points of view. In this brief, we focus on resource-efficient randomly connected neural networks known as random vector functional link (RVFL) networks since their simple design and extremely fast training time make them very attractive for solving many applied classification tasks. We propose to represent input features via the density-based encoding known in the area of stochastic computing and use the operations of binding and bundling from the area of hyperdimensional computing for obtaining the activations of the hidden neurons. Using a collection of 121 real-world data sets from the UCI machine learning repository, we empirically show that the proposed approach demonstrates higher average accuracy than the conventional RVFL. We also demonstrate that it is possible to represent the readout matrix using only integers in a limited range with minimal loss in the accuracy. In this case, the proposed approach operates only on small ${n}$ -bits integers, which results in a computationally efficient architecture. Finally, through hardware field-programmable gate array (FPGA) implementations, we show that such an approach consumes approximately 11 times less energy than that of the conventional RVFL.

Place, publisher, year, edition, pages
IEEE, 2021. Vol. 32, no 8, p. 3777-3783, article id 9174774
Keywords [en]
Density-based encoding, hyperdimensional computing, random vector functional link (RVFL) networks
National Category
Computer Sciences
Identifiers
URN: urn:nbn:se:umu:diva-186635DOI: 10.1109/TNNLS.2020.3015971ISI: 000681169500047PubMedID: 32833655Scopus ID: 2-s2.0-85112022593OAI: oai:DiVA.org:umu-186635DiVA, id: diva2:1589206
Funder
Swedish Research Council, 2015-04677Available from: 2021-08-30 Created: 2021-08-30 Last updated: 2022-05-10Bibliographically approved

Open Access in DiVA

No full text in DiVA

Other links

Publisher's full textPubMedScopusFree full text in ArXiv

Authority records

Wiklund, Urban

Search in DiVA

By author/editor
Wiklund, Urban
By organisation
Radiation Physics
In the same journal
IEEE Transactions on Neural Networks and Learning Systems
Computer Sciences

Search outside of DiVA

GoogleGoogle Scholar

doi
pubmed
urn-nbn

Altmetric score

doi
pubmed
urn-nbn
Total: 133 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf