mdh.sePublikasjoner
Endre søk
RefereraExporteraLink to record
Permanent link

Direct link
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annet format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annet språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf
ADONN: Adaptive design of optimized deep neural networks for embedded systems
Mälardalens högskola, Akademin för innovation, design och teknik, Inbyggda system.
Mälardalens högskola, Akademin för innovation, design och teknik, Inbyggda system.
Mälardalens högskola, Akademin för innovation, design och teknik, Inbyggda system.ORCID-id: 0000-0001-7586-0409
2018 (engelsk)Inngår i: Proceedings - 21st Euromicro Conference on Digital System Design, DSD 2018, Institute of Electrical and Electronics Engineers Inc. , 2018, s. 397-404Konferansepaper, Publicerat paper (Fagfellevurdert)
Abstract [en]

Nowadays, many modern applications, e.g. autonomous system, and cloud data services need to capture and process a big amount of raw data at runtime that ultimately necessitates a high-performance computing model. Deep Neural Network (DNN) has already revealed its learning capabilities in runtime data processing for modern applications. However, DNNs are becoming more deep sophisticated models for gaining higher accuracy which require a remarkable computing capacity. Considering high-performance cloud infrastructure as a supplier of required computational throughput is often not feasible. Instead, we intend to find a near-sensor processing solution which will lower the need for network bandwidth and increase privacy and power efficiency, as well as guaranteeing worst-case response-times. Toward this goal, we introduce ADONN framework, which aims to automatically design a highly robust DNN architecture for embedded devices as the closest processing unit to the sensors. ADONN adroitly searches the design space to find improved neural architectures. Our proposed framework takes advantage of a multi-objective evolutionary approach, which exploits a pruned design space inspired by a dense architecture. Unlike recent works that mainly have tried to generate highly accurate networks, ADONN also considers the network size factor as the second objective to build a highly optimized network fitting with limited computational resource budgets while delivers comparable accuracy level. In comparison with the best result on CIFAR-10 dataset, a generated network by ADONN presents up to 26.4 compression rate while loses only 4% accuracy. In addition, ADONN maps the generated DNN on the commodity programmable devices including ARM Processor, High-Performance CPU, GPU, and FPGA.

sted, utgiver, år, opplag, sider
Institute of Electrical and Electronics Engineers Inc. , 2018. s. 397-404
Emneord [en]
Approximation Computing, Multi-Objective Optimization, Neural Architectural Search, Neural Processing Unit, Budget control, Data handling, Distributed computer systems, Embedded systems, Evolutionary algorithms, Integrated circuit design, Multiobjective optimization, Network architecture, Neural networks, Systems analysis, Cloud infrastructures, Computational resources, High-performance computing models, Learning capabilities, Multi-objective evolutionary, Neural-processing, Deep neural networks
HSV kategori
Identifikatorer
URN: urn:nbn:se:mdh:diva-41390DOI: 10.1109/DSD.2018.00074Scopus ID: 2-s2.0-85056450132ISBN: 9781538673768 (tryckt)OAI: oai:DiVA.org:mdh-41390DiVA, id: diva2:1362220
Konferanse
21st Euromicro Conference on Digital System Design, DSD 2018, 29 August 2018 through 31 August 2018
Tilgjengelig fra: 2019-10-18 Laget: 2019-10-18 Sist oppdatert: 2019-12-18bibliografisk kontrollert

Open Access i DiVA

Fulltekst mangler i DiVA

Andre lenker

Forlagets fulltekstScopus

Personposter BETA

Loni, MohammadDaneshtalab, MasoudSjödin, Mikael

Søk i DiVA

Av forfatter/redaktør
Loni, MohammadDaneshtalab, MasoudSjödin, Mikael
Av organisasjonen

Søk utenfor DiVA

GoogleGoogle Scholar

doi
isbn
urn-nbn

Altmetric

doi
isbn
urn-nbn
RefereraExporteraLink to record
Permanent link

Direct link
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annet format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annet språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf