When Single Event Upset Meets Deep Neural Networks: Observations, Explorations, and Remedies
Deep Neural Network has proved its potential in various perception tasks and hence become an appealing option for interpretation and data processing in security sensitive systems. However, security-sensitive systems demand not only high perception performance, but also design robustness under variou...
Saved in:
| Published in: | Proceedings of the ASP-DAC ... Asia and South Pacific Design Automation Conference pp. 163 - 168 |
|---|---|
| Main Authors: | , , , , , |
| Format: | Conference Proceeding |
| Language: | English |
| Published: |
IEEE
01.01.2020
|
| Subjects: | |
| ISSN: | 2153-697X |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Abstract | Deep Neural Network has proved its potential in various perception tasks and hence become an appealing option for interpretation and data processing in security sensitive systems. However, security-sensitive systems demand not only high perception performance, but also design robustness under various circumstances. Unlike prior works that study network robustness from software level, we investigate from hardware perspective about the impact of Single Event Upset (SEU) induced parameter perturbation (SIPP) on neural networks. We systematically define the fault models of SEU and then provide the definition of sensitivity to SIPP as the robustness measure for the network. We are then able to analytically explore the weakness of a network and summarize the key findings for the impact of SIPP on different types of bits in a floating point parameter, layer-wise robustness within the same network and impact of network depth. Based on those findings, we propose two remedy solutions to protect DNNs from SIPPs, which can mitigate accuracy degradation from 28% to 0.27% for ResNet with merely 0.24-bit SRAM area overhead per parameter. |
|---|---|
| AbstractList | Deep Neural Network has proved its potential in various perception tasks and hence become an appealing option for interpretation and data processing in security sensitive systems. However, security-sensitive systems demand not only high perception performance, but also design robustness under various circumstances. Unlike prior works that study network robustness from software level, we investigate from hardware perspective about the impact of Single Event Upset (SEU) induced parameter perturbation (SIPP) on neural networks. We systematically define the fault models of SEU and then provide the definition of sensitivity to SIPP as the robustness measure for the network. We are then able to analytically explore the weakness of a network and summarize the key findings for the impact of SIPP on different types of bits in a floating point parameter, layer-wise robustness within the same network and impact of network depth. Based on those findings, we propose two remedy solutions to protect DNNs from SIPPs, which can mitigate accuracy degradation from 28% to 0.27% for ResNet with merely 0.24-bit SRAM area overhead per parameter. |
| Author | Zhuo, Cheng Hashimoto, Masanori Yan, Zheyu Shi, Yiyu Liao, Wang Zhou, Xichuan |
| Author_xml | – sequence: 1 givenname: Zheyu surname: Yan fullname: Yan, Zheyu organization: Zhejiang University – sequence: 2 givenname: Yiyu surname: Shi fullname: Shi, Yiyu organization: University of Notre Dame – sequence: 3 givenname: Wang surname: Liao fullname: Liao, Wang organization: Osaka University – sequence: 4 givenname: Masanori surname: Hashimoto fullname: Hashimoto, Masanori organization: Osaka University – sequence: 5 givenname: Xichuan surname: Zhou fullname: Zhou, Xichuan organization: Chongqing University – sequence: 6 givenname: Cheng surname: Zhuo fullname: Zhuo, Cheng organization: Zhejiang University |
| BookMark | eNo1kM1OwkAYRUejiYA8gZtxb_Gbv07HXQP4k6AYkejChEynX7Vapk2nor69JOLZnJzNXdw-OfC1R0JOGYwYA3OeLu6jSTqWWqt4xIHDyIBUTMg90meaJ0wyLmCf9DhTIoqNfj4iwxDeYYsCrhn0yMvTG3q6KP1rhXS6Qd_RZROwo7eIXaATxIbe4Wdrq626r7r9CBd0ngVsN7Yrax_O6PS7qer2v6zP6QOuMS8xHJPDwlYBhzsPyPJy-ji-jmbzq5txOossN6qLXMwLB5lBFxdCCSNMkRmtQHKRQxwnoE3hEJV2RlruUABYqV1iTQKFM7kYkJO_3RIRV01brm37s9qdIX4BjMhWug |
| ContentType | Conference Proceeding |
| DBID | 6IE 6IL CBEJK RIE RIL |
| DOI | 10.1109/ASP-DAC47756.2020.9045134 |
| DatabaseName | IEEE Electronic Library (IEL) Conference Proceedings IEEE Proceedings Order Plan All Online (POP All Online) 1998-present by volume IEEE Xplore All Conference Proceedings IEEE Electronic Library (IEL) IEEE Proceedings Order Plans (POP All) 1998-Present |
| DatabaseTitleList | |
| Database_xml | – sequence: 1 dbid: RIE name: IEEE Electronic Library (IEL) url: https://ieeexplore.ieee.org/ sourceTypes: Publisher |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Engineering Computer Science |
| EISBN | 1728141230 9781728141237 |
| EISSN | 2153-697X |
| EndPage | 168 |
| ExternalDocumentID | 9045134 |
| Genre | orig-research |
| GroupedDBID | 5VS 6IE 6IF 6IL 6IN AAWTH ABLEC ACGFS ADZIZ ALMA_UNASSIGNED_HOLDINGS APO AVWKF BEFXN BFFAM BGNUA BKEBE BPEOZ CBEJK CHZPO I07 IEGSK M43 OCL RIE RIL |
| ID | FETCH-LOGICAL-a295t-c62fc0b9ec6f353939fb9750423d0668079fcee57c94a2ce300a47c8a980fc9d3 |
| IEDL.DBID | RIE |
| ISICitedReferencesCount | 45 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001235809500032&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| IngestDate | Wed Aug 27 05:55:37 EDT 2025 |
| IsPeerReviewed | false |
| IsScholarly | false |
| Language | English |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-a295t-c62fc0b9ec6f353939fb9750423d0668079fcee57c94a2ce300a47c8a980fc9d3 |
| PageCount | 6 |
| ParticipantIDs | ieee_primary_9045134 |
| PublicationCentury | 2000 |
| PublicationDate | 2020-Jan. |
| PublicationDateYYYYMMDD | 2020-01-01 |
| PublicationDate_xml | – month: 01 year: 2020 text: 2020-Jan. |
| PublicationDecade | 2020 |
| PublicationTitle | Proceedings of the ASP-DAC ... Asia and South Pacific Design Automation Conference |
| PublicationTitleAbbrev | ASP-DAC |
| PublicationYear | 2020 |
| Publisher | IEEE |
| Publisher_xml | – name: IEEE |
| SSID | ssj0000502710 ssib055574204 |
| Score | 1.9843811 |
| Snippet | Deep Neural Network has proved its potential in various perception tasks and hence become an appealing option for interpretation and data processing in... |
| SourceID | ieee |
| SourceType | Publisher |
| StartPage | 163 |
| SubjectTerms | Hardware Neural networks Robustness Sensitivity Single event upsets Software Task analysis |
| Title | When Single Event Upset Meets Deep Neural Networks: Observations, Explorations, and Remedies |
| URI | https://ieeexplore.ieee.org/document/9045134 |
| WOSCitedRecordID | wos001235809500032&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3dT8IwEL8AMUZfUMD4nZr4yGCsLV19I3zEB0UikvBgQrrulpiQQdjw77fdBmrii0-79WHp7rre7vq73wHcR57sBEopx4RegcNEiI4fUOUo5lPjXrqhn1PmP4nx2J_P5aQEzX0tDCJm4DNsWTE7yw9XemtTZW1pyVAoK0NZiG5eq7VbO5xzE-QVoUXO620Cro57CHcFrWa7N504g16fCcEtNsFzW8XzfjVWyfzKqPq_GZ1A47tAj0z2rucUShjXoLrr0ECKD7YGxz_oBuvwbnbemEyNvEQytEBHMlsnmJJnxDQhA8Q1sWQdamkuGTo8eSAvwT5vmzRJDtnb3ak4JK9oj-cxacBsNHzrPzpFdwVHeZKnju56kXYDibobUU4llVEgLdm7R0PzH-K7QkbmNbjQkilPI3VdxYT2lfTdSMuQnkElXsV4DkRzY-2OiiRnIesoFQRaR6FS2vNRe5JdQN2qbrHOCTQWhdYu_x6-giNrnTzPcQ2VdLPFGzjQn-lHsrnNrP4FvmWtJQ |
| linkProvider | IEEE |
| linkToHtml | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1JT8JAFH7BJS4XFzDujolHCmUW2vFGEKMRkAgkHEzIdPqamJhKaPH3O9MW1MSLp1kOk9nfvDff-x7ATURlI1BKOUb1Chzuhej4AVOO4j4z4qUZ-jllftfr9_3JRA5KUF35wiBiBj7Dms1mf_nhh15YU1ldWjIUxtdgQ3BO3dxba7l7hBBGzSuUi5zZ26hcDXcLrgtizXprOHDuWm3uecKiE6hbK1r8FVolkyz3e__r0z5Uvl30yGAlfA6ghPEh7C1jNJDiyB7C7g_CwTK8mrs3JkOTf0fSsVBHMp4lmJIeYpqQO8QZsXQd6t0kGT48uSXPwcpym1RJDtpbllQckhe0H_SYVGB83xm1H5wivoKjqBSpo5s00m4gUTcjJphkMgqkpXunLDQvEd_1ZGSGITwtuaIamesq7mlfSd-NtAzZEazHHzEeA9HCrHdDRVLwkDeUCgKto1ApTX3UVPITKNupm85yCo1pMWunf1dfwfbDqNeddh_7T2ewY1cqt3qcw3o6X-AFbOrP9C2ZX2Y74AvtmrBs |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=proceeding&rft.title=Proceedings+of+the+ASP-DAC+...+Asia+and+South+Pacific+Design+Automation+Conference&rft.atitle=When+Single+Event+Upset+Meets+Deep+Neural+Networks%3A+Observations%2C+Explorations%2C+and+Remedies&rft.au=Yan%2C+Zheyu&rft.au=Shi%2C+Yiyu&rft.au=Liao%2C+Wang&rft.au=Hashimoto%2C+Masanori&rft.date=2020-01-01&rft.pub=IEEE&rft.eissn=2153-697X&rft.spage=163&rft.epage=168&rft_id=info:doi/10.1109%2FASP-DAC47756.2020.9045134&rft.externalDocID=9045134 |