elib
DLR-Header
DLR-Logo -> http://www.dlr.de
DLR Portal Home | Impressum | Datenschutz | Kontakt | English
Schriftgröße: [-] Text [+]

Resilient training of neural network classifiers with approximate computing techniques for hardware-optimised implementations

Ferreira Torres, Vitor und Sill Torres, Frank (2019) Resilient training of neural network classifiers with approximate computing techniques for hardware-optimised implementations. IET Computers and Digital Techniques. Institution of Engineering and Technology (IET). doi: 10.1049/iet-cdt.2019.0036. ISSN 1751-8601.

Dieses Archiv kann nicht den Volltext zur Verfügung stellen.

Offizielle URL: https://digital-library.theiet.org/content/journals/10.1049/iet-cdt.2019.0036

Kurzfassung

As Machine Learning applications increase the demand for optimised implementations in both embedded and high-end processing platforms, the industry and research community have been responding with different approaches to implement these solutions. This work presents approximations to arithmetic operations and mathematical functions that, associated with a customised adaptive artificial neural networks training method, based on RMSProp, provide reliable and efficient implementations of classifiers. The proposed solution does not rely on mixed operations with higher precision or complex rounding methods that are commonly applied. The intention of this work is not to find the optimal simplifications for specific deep learning problems but to present an optimised framework that can be used as reliably as one implemented with precise operations, standard training algorithms and the same network structures and hyper-parameters. By simplifying the 'half-precision' floating point format and approximating exponentiation and square root operations, the authors' work drastically reduces the field programmable gate array implementation complexity (e.g. -43 and -57% in two of the component resources). The reciprocal square root approximation is so simple it could be implemented only with combination logic. In a full software implementation for a mixed-precision platform, only two of the approximations compensate the processing overhead of precision conversions.

elib-URL des Eintrags:https://elib.dlr.de/129585/
Dokumentart:Zeitschriftenbeitrag
Titel:Resilient training of neural network classifiers with approximate computing techniques for hardware-optimised implementations
Autoren:
AutorenInstitution oder E-Mail-AdresseAutoren-ORCID-iDORCID Put Code
Ferreira Torres, VitorNICHT SPEZIFIZIERTNICHT SPEZIFIZIERTNICHT SPEZIFIZIERT
Sill Torres, FrankFrank.SillTorres (at) dlr.dehttps://orcid.org/0000-0002-4028-455XNICHT SPEZIFIZIERT
Datum:September 2019
Erschienen in:IET Computers and Digital Techniques
Referierte Publikation:Ja
Open Access:Nein
Gold Open Access:Nein
In SCOPUS:Ja
In ISI Web of Science:Ja
DOI:10.1049/iet-cdt.2019.0036
Verlag:Institution of Engineering and Technology (IET)
ISSN:1751-8601
Status:veröffentlicht
Stichwörter:Machine Learning, Approximations
HGF - Forschungsbereich:keine Zuordnung
HGF - Programm:keine Zuordnung
HGF - Programmthema:keine Zuordnung
DLR - Schwerpunkt:keine Zuordnung
DLR - Forschungsgebiet:keine Zuordnung
DLR - Teilgebiet (Projekt, Vorhaben):keine Zuordnung
Standort: Bremerhaven
Institute & Einrichtungen:Institut für den Schutz maritimer Infrastrukturen > Resilienz Maritimer Systeme
Hinterlegt von: Sill Torres, Dr. Frank
Hinterlegt am:28 Okt 2019 13:19
Letzte Änderung:19 Nov 2021 20:39

Nur für Mitarbeiter des Archivs: Kontrollseite des Eintrags

Blättern
Suchen
Hilfe & Kontakt
Informationen
electronic library verwendet EPrints 3.3.12
Gestaltung Webseite und Datenbank: Copyright © Deutsches Zentrum für Luft- und Raumfahrt (DLR). Alle Rechte vorbehalten.