Denoised Internal Models: a Brain-Inspired Autoencoder against Adversarial Attacks

Despite its great success, deep learning severely suffers from robustness; that is, deep neural networks are very vulnerable to adversarial attacks, even the simplest ones. Inspired by recent advances in brain science, we propose the Denoised Internal Models (DIM), a novel generative autoencoder-bas...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:arXiv.org
Hlavní autoři: Liu, Kaiyuan, Li, Xingyu, Lai, Yurui, Zhang, Ge, Su, Hang, Wang, Jiachen, Guo, Chunxu, Guan, Jisong, Zhou, Yi
Médium: Paper
Jazyk:angličtina
Vydáno: Ithaca Cornell University Library, arXiv.org 05.03.2023
Témata:
ISSN:2331-8422
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Abstract Despite its great success, deep learning severely suffers from robustness; that is, deep neural networks are very vulnerable to adversarial attacks, even the simplest ones. Inspired by recent advances in brain science, we propose the Denoised Internal Models (DIM), a novel generative autoencoder-based model to tackle this challenge. Simulating the pipeline in the human brain for visual signal processing, DIM adopts a two-stage approach. In the first stage, DIM uses a denoiser to reduce the noise and the dimensions of inputs, reflecting the information pre-processing in the thalamus. Inspired from the sparse coding of memory-related traces in the primary visual cortex, the second stage produces a set of internal models, one for each category. We evaluate DIM over 42 adversarial attacks, showing that DIM effectively defenses against all the attacks and outperforms the SOTA on the overall robustness.
AbstractList Despite its great success, deep learning severely suffers from robustness; that is, deep neural networks are very vulnerable to adversarial attacks, even the simplest ones. Inspired by recent advances in brain science, we propose the Denoised Internal Models (DIM), a novel generative autoencoder-based model to tackle this challenge. Simulating the pipeline in the human brain for visual signal processing, DIM adopts a two-stage approach. In the first stage, DIM uses a denoiser to reduce the noise and the dimensions of inputs, reflecting the information pre-processing in the thalamus. Inspired from the sparse coding of memory-related traces in the primary visual cortex, the second stage produces a set of internal models, one for each category. We evaluate DIM over 42 adversarial attacks, showing that DIM effectively defenses against all the attacks and outperforms the SOTA on the overall robustness.
Author Lai, Yurui
Zhang, Ge
Guan, Jisong
Li, Xingyu
Wang, Jiachen
Zhou, Yi
Guo, Chunxu
Liu, Kaiyuan
Su, Hang
Author_xml – sequence: 1
  givenname: Kaiyuan
  surname: Liu
  fullname: Liu, Kaiyuan
– sequence: 2
  givenname: Xingyu
  surname: Li
  fullname: Li, Xingyu
– sequence: 3
  givenname: Yurui
  surname: Lai
  fullname: Lai, Yurui
– sequence: 4
  givenname: Ge
  surname: Zhang
  fullname: Zhang, Ge
– sequence: 5
  givenname: Hang
  surname: Su
  fullname: Su, Hang
– sequence: 6
  givenname: Jiachen
  surname: Wang
  fullname: Wang, Jiachen
– sequence: 7
  givenname: Chunxu
  surname: Guo
  fullname: Guo, Chunxu
– sequence: 8
  givenname: Jisong
  surname: Guan
  fullname: Guan, Jisong
– sequence: 9
  givenname: Yi
  surname: Zhou
  fullname: Zhou, Yi
BookMark eNotj8FKw0AURQdRsNZ-gLuA69R5b_KSqbtYqwYqgnRfJjMvkhomdSYpfr4RXd3FuefCvRLnvvcsxA3IZaaJ5J0J3-1piQCwBKmz7EzMUClIdYZ4KRYxHqSUmBdIpGbi_ZF930Z2SeUHDt50yWvvuIv3iUkegml9Wvl4bMPUKMehZ28nHBLzMaE4JKU7cYgmtJNYDoOxn_FaXDSmi7z4z7nYPW1265d0-_Zcrcttagh1WueMrsYGwGhHiEgOVEM12oydlZYKJ61SetVQoRtHZMmYjNESsAIo1Fzc_s0eQ_81chz2h378PRD3mEsAWhWo1Q-XfFMF
ContentType Paper
Copyright 2023. This work is published under http://arxiv.org/licenses/nonexclusive-distrib/1.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Copyright_xml – notice: 2023. This work is published under http://arxiv.org/licenses/nonexclusive-distrib/1.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
DBID 8FE
8FG
ABJCF
ABUWG
AFKRA
AZQEC
BENPR
BGLVJ
CCPQU
DWQXO
HCIFZ
L6V
M7S
PHGZM
PHGZT
PIMPY
PKEHL
PQEST
PQGLB
PQQKQ
PQUKI
PRINS
PTHSS
DOI 10.48550/arxiv.2111.10844
DatabaseName ProQuest SciTech Collection
ProQuest Technology Collection
ProQuest SciTech Premium Collection Technology Collection Materials Science & Engineering Database
ProQuest Central (Alumni)
ProQuest Central UK/Ireland
ProQuest Central Essentials - QC
ProQuest Central
Technology Collection
ProQuest One
ProQuest Central Korea
SciTech Premium Collection
ProQuest Engineering Collection
Engineering Database
ProQuest Central Premium
ProQuest One Academic (New)
ProQuest Publicly Available Content Database
ProQuest One Academic Middle East (New)
ProQuest One Academic Eastern Edition (DO NOT USE)
ProQuest One Applied & Life Sciences
ProQuest One Academic (retired)
ProQuest One Academic UKI Edition
ProQuest Central China
Engineering Collection
DatabaseTitle Publicly Available Content Database
Engineering Database
Technology Collection
ProQuest One Academic Middle East (New)
ProQuest Central Essentials
ProQuest One Academic Eastern Edition
ProQuest Central (Alumni Edition)
SciTech Premium Collection
ProQuest One Community College
ProQuest Technology Collection
ProQuest SciTech Collection
ProQuest Central China
ProQuest Central
ProQuest One Applied & Life Sciences
ProQuest Engineering Collection
ProQuest One Academic UKI Edition
ProQuest Central Korea
Materials Science & Engineering Collection
ProQuest Central (New)
ProQuest One Academic
ProQuest One Academic (New)
Engineering Collection
DatabaseTitleList Publicly Available Content Database
Database_xml – sequence: 1
  dbid: PIMPY
  name: ProQuest Publicly Available Content Database
  url: http://search.proquest.com/publiccontent
  sourceTypes: Aggregation Database
DeliveryMethod fulltext_linktorsrc
Discipline Physics
EISSN 2331-8422
Genre Working Paper/Pre-Print
GroupedDBID 8FE
8FG
ABJCF
ABUWG
AFKRA
ALMA_UNASSIGNED_HOLDINGS
AZQEC
BENPR
BGLVJ
CCPQU
DWQXO
FRJ
HCIFZ
L6V
M7S
M~E
PHGZM
PHGZT
PIMPY
PKEHL
PQEST
PQGLB
PQQKQ
PQUKI
PRINS
PTHSS
ID FETCH-LOGICAL-a528-b6e2db2f11a8d52225d13f5b2c4edc0c57d0c3389f578fd55c5aa4e2c51e31173
IEDL.DBID BENPR
IngestDate Mon Jun 30 09:30:05 EDT 2025
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed false
IsScholarly false
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-a528-b6e2db2f11a8d52225d13f5b2c4edc0c57d0c3389f578fd55c5aa4e2c51e31173
Notes SourceType-Working Papers-1
ObjectType-Working Paper/Pre-Print-1
content type line 50
OpenAccessLink https://www.proquest.com/docview/2601159728?pq-origsite=%requestingapplication%
PQID 2601159728
PQPubID 2050157
ParticipantIDs proquest_journals_2601159728
PublicationCentury 2000
PublicationDate 20230305
PublicationDateYYYYMMDD 2023-03-05
PublicationDate_xml – month: 03
  year: 2023
  text: 20230305
  day: 05
PublicationDecade 2020
PublicationPlace Ithaca
PublicationPlace_xml – name: Ithaca
PublicationTitle arXiv.org
PublicationYear 2023
Publisher Cornell University Library, arXiv.org
Publisher_xml – name: Cornell University Library, arXiv.org
SSID ssj0002672553
Score 1.8245864
SecondaryResourceType preprint
Snippet Despite its great success, deep learning severely suffers from robustness; that is, deep neural networks are very vulnerable to adversarial attacks, even the...
SourceID proquest
SourceType Aggregation Database
SubjectTerms Artificial neural networks
Brain
Machine learning
Noise reduction
Robustness
Signal processing
Thalamus
Visual signals
Title Denoised Internal Models: a Brain-Inspired Autoencoder against Adversarial Attacks
URI https://www.proquest.com/docview/2601159728
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV07T8MwELagBYmJt3iUygOraR3HccKCWmhFB6qodCgTcvxAlaqkxGnFz8c2KQxILIyWPVi273y--_x9AFx3GeOSUIYk5QkKSZSgLIszlESCKmZDFC2FF5tg43E8myVpnXAzNaxy4xO9o5aFcDnyjqO-slcvC-K75TtyqlGuulpLaGyDpmMqs-e82R-M08l3liWImI2ZyVc505N3dXj5MV_f2OHY4evC8JcT9jfLcP-_czoAzZQvVXkItlR-BHY9olOYYzB5UHkxN0rCOum3gE73bGFuIYd9JwyBRrkrs9sRvVVVOEJLqUrI32yXqaBXajbcnU_Yqyr3Ff8ETIeD6f0jqgUUEKdBjLJIBTILNMY8ltQ97CQmmmaBCJUUXUGZ7Ar7RE20NVstKRWU81AFgmJFMGbkFDTyIldnAHZ55Lj6PHOUDQFUTAXXTGJNMiISrs9Ba7NCr7URmNef5bn4u_sS7DkVdw_toi3QqMqVugI7Yl3NTdmu97TtYJnPtpWOntKXT3LOr30
linkProvider ProQuest
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMw1V1LS8QwEB50VfTkG9_moMe4Tdo0rSCirmJRl0X2oKeSJqksSKttff0o_6NJbfUgePPgOaWQzGQyM_nyfQA7DudCuYxjxUSIPdcPcZIECQ59yTQ3KUqqZC02wfv94OYmHIzBe_sWxsIq25hYB2qVS9sj71rqK3P0chocPjxiqxplb1dbCY1Pt7jQby-mZCsPop6x7y6lZ6fDk3PcqApgwWiAE19TldCUEBEoZqsdRdyUJVR6WklHMq4caeq2MDW-nCrGJBPC01Qyol1CuGt-Ow4TnvF1pwMTg-hqcPvV1KE-Nym6-3l7WnOFdUXxOnreM2UWsXA-z_sR8-uD7Gz2ny3BnJm6eNDFPIzpbAGmaryqLBfhuqezfFRqhZqW5j2yqm735T4S6NjKXuAosyAC88XRU5Vbuk6lCyTuzFBZoVqHuhR296GjqrJEA0sw_ItZLEMnyzO9AsgRvmUirHmxTIKjAyZFyhVJ3cSVoUhXYaM1SNxs8TL-tsba78PbMH0-vLqML6P-xTrMWL36GsTGNqBTFU96EyblczUqi63GnRDEf2y9D83DCS8
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=article&rft.atitle=Denoised+Internal+Models%3A+a+Brain-Inspired+Autoencoder+against+Adversarial+Attacks&rft.jtitle=arXiv.org&rft.au=Liu%2C+Kaiyuan&rft.au=Li%2C+Xingyu&rft.au=Lai%2C+Yurui&rft.au=Zhang%2C+Ge&rft.date=2023-03-05&rft.pub=Cornell+University+Library%2C+arXiv.org&rft.eissn=2331-8422&rft_id=info:doi/10.48550%2Farxiv.2111.10844