Permutation Equivariance of Transformers and Its Applications
Revolutionizing the field of deep learning, Transformer-based models have achieved remarkable performance in many tasks. Recent research has recognized these models are robust to shuffling but are limited to inter-token permutation in the forward propagation. In this work, we propose our definition...
Saved in:
Published in | arXiv.org |
---|---|
Main Authors | , , , , , |
Format | Paper |
Language | English |
Published |
Ithaca
Cornell University Library, arXiv.org
31.03.2024
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Abstract | Revolutionizing the field of deep learning, Transformer-based models have achieved remarkable performance in many tasks. Recent research has recognized these models are robust to shuffling but are limited to inter-token permutation in the forward propagation. In this work, we propose our definition of permutation equivariance, a broader concept covering both inter- and intra- token permutation in the forward and backward propagation of neural networks. We rigorously proved that such permutation equivariance property can be satisfied on most vanilla Transformer-based models with almost no adaptation. We examine the property over a range of state-of-the-art models including ViT, Bert, GPT, and others, with experimental validations. Further, as a proof-of-concept, we explore how real-world applications including privacy-enhancing split learning, and model authorization, could exploit the permutation equivariance property, which implicates wider, intriguing application scenarios. |
---|---|
AbstractList | Revolutionizing the field of deep learning, Transformer-based models have achieved remarkable performance in many tasks. Recent research has recognized these models are robust to shuffling but are limited to inter-token permutation in the forward propagation. In this work, we propose our definition of permutation equivariance, a broader concept covering both inter- and intra- token permutation in the forward and backward propagation of neural networks. We rigorously proved that such permutation equivariance property can be satisfied on most vanilla Transformer-based models with almost no adaptation. We examine the property over a range of state-of-the-art models including ViT, Bert, GPT, and others, with experimental validations. Further, as a proof-of-concept, we explore how real-world applications including privacy-enhancing split learning, and model authorization, could exploit the permutation equivariance property, which implicates wider, intriguing application scenarios. |
Author | Chu, Pengzhi Ye, Hangyu Li, Baochun Liyao Xiang Xu, Hengyuan Yao, Dixi |
Author_xml | – sequence: 1 givenname: Hengyuan surname: Xu fullname: Xu, Hengyuan – sequence: 2 fullname: Liyao Xiang – sequence: 3 givenname: Hangyu surname: Ye fullname: Ye, Hangyu – sequence: 4 givenname: Dixi surname: Yao fullname: Yao, Dixi – sequence: 5 givenname: Pengzhi surname: Chu fullname: Chu, Pengzhi – sequence: 6 givenname: Baochun surname: Li fullname: Li, Baochun |
BookMark | eNqNjc0KgkAURoco6M93GGgt6J00Ny0ijNq1cC-DjTCid_TemZ4_iR6g1bc45_BtxRIdmoXYgFJpXBwB1iJi7pIkgfwEWaY24vw0NASvvXUoyynYtyarsTHStbIijdw6Ggyx1PiSD8_yMo69bb4B78Wq1T2b6Lc7cbiV1fUej-SmYNjXnQuEM6qhmE9zSKFQ_1kfAgQ6JA |
ContentType | Paper |
Copyright | 2024. 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: 2024. 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 PIMPY PQEST PQQKQ PQUKI PRINS PTHSS |
DatabaseName | ProQuest SciTech Collection ProQuest Technology Collection Materials Science & Engineering Collection ProQuest Central (Alumni) ProQuest Central ProQuest Central Essentials ProQuest Central Technology Collection ProQuest One Community College ProQuest Central SciTech Premium Collection ProQuest Engineering Collection Engineering Database Publicly Available Content Database ProQuest One Academic Eastern Edition (DO NOT USE) ProQuest One Academic ProQuest One Academic UKI Edition ProQuest Central China Engineering Collection |
DatabaseTitle | Publicly Available Content Database Engineering Database Technology Collection 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 Engineering Collection ProQuest One Academic UKI Edition ProQuest Central Korea Materials Science & Engineering Collection ProQuest One Academic Engineering Collection |
DatabaseTitleList | Publicly Available Content Database |
Database_xml | – sequence: 1 dbid: 8FG name: ProQuest Technology Collection url: https://search.proquest.com/technologycollection1 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 PIMPY PQEST PQQKQ PQUKI PRINS PTHSS |
ID | FETCH-proquest_journals_28026621283 |
IEDL.DBID | BENPR |
IngestDate | Thu Oct 10 20:20:01 EDT 2024 |
IsOpenAccess | true |
IsPeerReviewed | false |
IsScholarly | false |
Language | English |
LinkModel | DirectLink |
MergedId | FETCHMERGED-proquest_journals_28026621283 |
OpenAccessLink | https://www.proquest.com/docview/2802662128?pq-origsite=%requestingapplication% |
PQID | 2802662128 |
PQPubID | 2050157 |
ParticipantIDs | proquest_journals_2802662128 |
PublicationCentury | 2000 |
PublicationDate | 20240331 |
PublicationDateYYYYMMDD | 2024-03-31 |
PublicationDate_xml | – month: 03 year: 2024 text: 20240331 day: 31 |
PublicationDecade | 2020 |
PublicationPlace | Ithaca |
PublicationPlace_xml | – name: Ithaca |
PublicationTitle | arXiv.org |
PublicationYear | 2024 |
Publisher | Cornell University Library, arXiv.org |
Publisher_xml | – name: Cornell University Library, arXiv.org |
SSID | ssj0002672553 |
Score | 3.5378275 |
SecondaryResourceType | preprint |
Snippet | Revolutionizing the field of deep learning, Transformer-based models have achieved remarkable performance in many tasks. Recent research has recognized these... |
SourceID | proquest |
SourceType | Aggregation Database |
SubjectTerms | Coders Equivalence Learning Neural networks Permutations Privacy Training Transformers |
Title | Permutation Equivariance of Transformers and Its Applications |
URI | https://www.proquest.com/docview/2802662128 |
hasFullText | 1 |
inHoldings | 1 |
isFullTextHit | |
isPrint | |
link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwfV1LSwMxEB5sF8FbfeGjloBeg-wjyfYkVXatQkuRCr2VPMGDtt2HR3-7s9tdFYQeQyAhD76Zb75JBuBGOaNNyARlVioaCS2o5NxSrU3MhLPhsP6BbzLl49foecEWTcAtb9IqW0ysgdqsdBUjvw1iZAscgTa-W29oVTWqUlebEhod8AI_qmRa7z6Zzl5-oiwBF-gzh_-AtrYeaQ-8mVzb7BD27McR7NdJlzo_BsSj7L3cauEk2ZRvn0hcq1MgK0fmrUeJ_hlBuk-eipyM_sjNJ3CdJvOHMW3nXDb3Il_-riI8hS4SfHsGRMRchr5ijkciYs5KXwpfGWmcUXpo1Tn0d410sbv7Eg4CNMTbd3R96BZZaa_QkBZqAJ04fRw0e4atyVfyDaHSf-w |
link.rule.ids | 786,790,12792,21416,33406,33777,43633,43838 |
linkProvider | ProQuest |
linkToHtml | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwfV1ZS8NAEB40QfTNE4-qC_q6SI7dTZ9EJSXVNhSJ0LewJ_igbXP4-92kiQpCnxd22YPvm29mdgbgVhglVUAYJpoLHDLJMKdUYylVRJjRwbCtwDdNafIWPs_JvHO4lV1aZY-JLVCrhWx85Hd-ZNUCtUAb3S9XuOka1URXuxYa2-A2JTcjB9zHOJ29_nhZfMqszRz8A9qWPUb74M74UhcHsKU_D2GnTbqU5RFYPCo-6nUsHMWr-v3LCtfmFtDCoKy3KK19hqzcR-OqRA9_ws3HcDOKs6cE92vm3bso899dBCfgWIGvTwGxiPLAE8TQkIXEaO5x5gnFlVFCDrU4g8Gmmc43D1_DbpJNJ_lknL5cwJ5vSXn9p24ATlXU-tKSaiWuupP7BshJgMs |
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=Permutation+Equivariance+of+Transformers+and+Its+Applications&rft.jtitle=arXiv.org&rft.au=Xu%2C+Hengyuan&rft.au=Liyao+Xiang&rft.au=Ye%2C+Hangyu&rft.au=Yao%2C+Dixi&rft.date=2024-03-31&rft.pub=Cornell+University+Library%2C+arXiv.org&rft.eissn=2331-8422 |