Show simple item record

dc.contributor.authorHassan, Salma
dc.contributor.authorAttia, Sameh
dc.contributor.authorSalama, Khaled N.
dc.contributor.authorMostafa, Hassan
dc.date.accessioned2020-05-07T11:06:51Z
dc.date.available2020-05-07T11:06:51Z
dc.date.issued2020-05-04
dc.date.submitted2020-03-12
dc.identifier.citationHassan, S., Attia, S., Salama, K. N., & Mostafa, H. (2020). EANN: Energy Adaptive Neural Networks. Electronics, 9(5), 746. doi:10.3390/electronics9050746
dc.identifier.issn2079-9292
dc.identifier.doi10.3390/electronics9050746
dc.identifier.urihttp://hdl.handle.net/10754/662760
dc.description.abstractThis paper proposes an Energy Adaptive Feedforward Neural Network (EANN). It uses multiple approximation techniques in the hardware implementation of the neuron unit. The used techniques are precision scaling, approximate multiplier, computation skipping, neuron skipping, activation function approximation and truncated accumulation. The proposed EANN system applies the partial dynamic reconfiguration (PDR) feature supported by the FPGA platform to reconfigure the hardware elements of the neural network based on the energy budget. The PDR technique enables the EANN system to remain functioning when the available energy budget is reduced by factors of 46.2% to 79.8% of the total energy of the unapproximated neural network. Unlike the conventional operation that only uses certain amount of energy and cannot function properly if the energy budget falls below that energy level, the EANN system remains functioning for longer time after energy drop at the expense of less accuracy. The proposed EANN system is highly recommended in limited-energy applications as it adapts the hardware units to the degraded energy at the expense of some accuracy loss.
dc.description.sponsorshipThis work was partially funded by ONE Lab at Zewail City of Science and Technology and at Cairo University, NTRA, ITIDA, ASRT, and NSERC.
dc.publisherMDPI AG
dc.relation.urlhttps://www.mdpi.com/2079-9292/9/5/746
dc.relation.urlhttps://www.mdpi.com/2079-9292/9/5/746/pdf
dc.relation.urlhttps://www.mdpi.com/2079-9292/9/5/746/pdf
dc.rightsArchived with thanks to Electronics
dc.rightsThis file is an open access version redistributed from: https://www.mdpi.com/2079-9292/9/5/746/pdf
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.titleEANN: Energy Adaptive Neural Networks
dc.typeArticle
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
dc.contributor.departmentElectrical Engineering Program
dc.contributor.departmentSensors Lab
dc.identifier.journalElectronics
dc.eprint.versionPublisher's Version/PDF
dc.contributor.institutionDepartment of Electronics and Communications Engineering, Cairo University, Giza 12613, Egypt.
dc.contributor.institutionUniversity of Science and technology, Nanotechnology and Nanoelectronics Program, Zewail City of Science and Technology, October Gardens, 6th of October, Giza 12578, Egypt.
dc.identifier.volume9
dc.identifier.issue5
dc.identifier.pages746
kaust.personSalama, Khaled N.
dc.date.accepted2020-04-29
refterms.dateFOA2020-05-07T11:07:57Z


Files in this item

Thumbnail
Name:
Articlefile1.pdf
Size:
1.451Mb
Format:
PDF
Description:
Publisher's Version/PDF

This item appears in the following Collection(s)

Show simple item record

Archived with thanks to Electronics
Except where otherwise noted, this item's license is described as Archived with thanks to Electronics