No Cover Image

Conference Paper/Proceeding/Abstract 240 views 4 downloads

Gradients Stand-in for Defending Deep Leakage in Federated Learning

Freya Hu, Hans Ren, Chen Hu, Yiming Li, Jingjing Deng, Xianghua Xie Orcid Logo

International Conference on Computing in Natural Sciences, Biomedicine and Engineering

Swansea University Authors: Freya Hu, Hans Ren, Chen Hu, Yiming Li, Xianghua Xie Orcid Logo

  • 66608.pdf

    PDF | Accepted Manuscript

    Author accepted manuscript document released under the terms of a Creative Commons CC-BY licence using the Swansea University Research Publications Policy (rights retention).

    Download (1.99MB)

Abstract

Federated Learning (FL) has become a cornerstone of privacy protection, shifting the paradigm towards localizing sensitive data while only sending model gradients to a central server. This strategy is designed to reinforce privacy protections and minimize the vulnerabilities inherent in centralized...

Full description

Published in: International Conference on Computing in Natural Sciences, Biomedicine and Engineering
Published: 2024
URI: https://cronfa.swan.ac.uk/Record/cronfa66608
Tags: Add Tag
No Tags, Be the first to tag this record!
first_indexed 2024-06-07T13:11:18Z
last_indexed 2024-06-07T13:11:18Z
id cronfa66608
recordtype SURis
fullrecord <?xml version="1.0" encoding="utf-8"?><rfc1807 xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:xsd="http://www.w3.org/2001/XMLSchema"><bib-version>v2</bib-version><id>66608</id><entry>2024-06-07</entry><title>Gradients Stand-in for Defending Deep Leakage in Federated Learning</title><swanseaauthors><author><sid>aa73524c5e3969c88fb7a3a5bde919b1</sid><firstname>Freya</firstname><surname>Hu</surname><name>Freya Hu</name><active>true</active><ethesisStudent>false</ethesisStudent></author><author><sid>9e043b899a2b786672a28ed4f864ffcc</sid><firstname>Hans</firstname><surname>Ren</surname><name>Hans Ren</name><active>true</active><ethesisStudent>false</ethesisStudent></author><author><sid>55d3ba5f8378c2e3439d7e3962aee726</sid><firstname>Chen</firstname><surname>Hu</surname><name>Chen Hu</name><active>true</active><ethesisStudent>false</ethesisStudent></author><author><sid>1b3389c4ef5d90bedb2a23843041ed68</sid><firstname>Yiming</firstname><surname>Li</surname><name>Yiming Li</name><active>true</active><ethesisStudent>false</ethesisStudent></author><author><sid>b334d40963c7a2f435f06d2c26c74e11</sid><ORCID>0000-0002-2701-8660</ORCID><firstname>Xianghua</firstname><surname>Xie</surname><name>Xianghua Xie</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2024-06-07</date><deptcode>MACS</deptcode><abstract>Federated Learning (FL) has become a cornerstone of privacy protection, shifting the paradigm towards localizing sensitive data while only sending model gradients to a central server. This strategy is designed to reinforce privacy protections and minimize the vulnerabilities inherent in centralized data storage systems. Despite its innovative approach, recent empirical studies have highlighted potential weaknesses in FL, notably regarding the exchange of gradients. In response, this studyintroduces a novel, efficacious method aimed at safeguarding against gradient leakage, namely, “AdaDefense”. Following the idea that model convergence can be achieved by using differenttypes of optimization methods, we suggest using a local standin rather than the actual local gradient for global gradient aggregation on the central server. This proposed approach not only effectively prevents gradient leakage, but also ensures that the overall performance of the model remains largely unaffected. Delving into the theoretical dimensions, we explore how gradients may inadvertently leak private information and present a theoretical framework supporting the efficacy of our proposed method. Extensive empirical tests, supported by popular benchmark experiments,validate that our approach maintains model integrity and is robust against gradient leakage, marking an important step in our pursuit of safe and efficient FL.</abstract><type>Conference Paper/Proceeding/Abstract</type><journal>International Conference on Computing in Natural Sciences, Biomedicine and Engineering</journal><volume/><journalNumber/><paginationStart/><paginationEnd/><publisher/><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint/><issnElectronic/><keywords/><publishedDay>10</publishedDay><publishedMonth>9</publishedMonth><publishedYear>2024</publishedYear><publishedDate>2024-09-10</publishedDate><doi/><url/><notes>In Press</notes><college>COLLEGE NANME</college><department>Mathematics and Computer Science School</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>MACS</DepartmentCode><institution>Swansea University</institution><apcterm/><funders/><projectreference/><lastEdited>2024-10-30T15:53:48.6845571</lastEdited><Created>2024-06-07T14:05:03.2631412</Created><path><level id="1">Faculty of Science and Engineering</level><level id="2">School of Mathematics and Computer Science - Computer Science</level></path><authors><author><firstname>Freya</firstname><surname>Hu</surname><order>1</order></author><author><firstname>Hans</firstname><surname>Ren</surname><order>2</order></author><author><firstname>Chen</firstname><surname>Hu</surname><order>3</order></author><author><firstname>Yiming</firstname><surname>Li</surname><order>4</order></author><author><firstname>Jingjing</firstname><surname>Deng</surname><order>5</order></author><author><firstname>Xianghua</firstname><surname>Xie</surname><orcid>0000-0002-2701-8660</orcid><order>6</order></author></authors><documents><document><filename>66608__30569__e265338858014f73b847482cc7c91c7f.pdf</filename><originalFilename>66608.pdf</originalFilename><uploaded>2024-06-07T14:10:46.7788234</uploaded><type>Output</type><contentLength>2091549</contentLength><contentType>application/pdf</contentType><version>Accepted Manuscript</version><cronfaStatus>true</cronfaStatus><documentNotes>Author accepted manuscript document released under the terms of a Creative Commons CC-BY licence using the Swansea University Research Publications Policy (rights retention).</documentNotes><copyrightCorrect>true</copyrightCorrect><language>eng</language><licence>https://creativecommons.org/licenses/by/4.0/deed.en</licence></document></documents><OutputDurs/></rfc1807>
spelling v2 66608 2024-06-07 Gradients Stand-in for Defending Deep Leakage in Federated Learning aa73524c5e3969c88fb7a3a5bde919b1 Freya Hu Freya Hu true false 9e043b899a2b786672a28ed4f864ffcc Hans Ren Hans Ren true false 55d3ba5f8378c2e3439d7e3962aee726 Chen Hu Chen Hu true false 1b3389c4ef5d90bedb2a23843041ed68 Yiming Li Yiming Li true false b334d40963c7a2f435f06d2c26c74e11 0000-0002-2701-8660 Xianghua Xie Xianghua Xie true false 2024-06-07 MACS Federated Learning (FL) has become a cornerstone of privacy protection, shifting the paradigm towards localizing sensitive data while only sending model gradients to a central server. This strategy is designed to reinforce privacy protections and minimize the vulnerabilities inherent in centralized data storage systems. Despite its innovative approach, recent empirical studies have highlighted potential weaknesses in FL, notably regarding the exchange of gradients. In response, this studyintroduces a novel, efficacious method aimed at safeguarding against gradient leakage, namely, “AdaDefense”. Following the idea that model convergence can be achieved by using differenttypes of optimization methods, we suggest using a local standin rather than the actual local gradient for global gradient aggregation on the central server. This proposed approach not only effectively prevents gradient leakage, but also ensures that the overall performance of the model remains largely unaffected. Delving into the theoretical dimensions, we explore how gradients may inadvertently leak private information and present a theoretical framework supporting the efficacy of our proposed method. Extensive empirical tests, supported by popular benchmark experiments,validate that our approach maintains model integrity and is robust against gradient leakage, marking an important step in our pursuit of safe and efficient FL. Conference Paper/Proceeding/Abstract International Conference on Computing in Natural Sciences, Biomedicine and Engineering 10 9 2024 2024-09-10 In Press COLLEGE NANME Mathematics and Computer Science School COLLEGE CODE MACS Swansea University 2024-10-30T15:53:48.6845571 2024-06-07T14:05:03.2631412 Faculty of Science and Engineering School of Mathematics and Computer Science - Computer Science Freya Hu 1 Hans Ren 2 Chen Hu 3 Yiming Li 4 Jingjing Deng 5 Xianghua Xie 0000-0002-2701-8660 6 66608__30569__e265338858014f73b847482cc7c91c7f.pdf 66608.pdf 2024-06-07T14:10:46.7788234 Output 2091549 application/pdf Accepted Manuscript true Author accepted manuscript document released under the terms of a Creative Commons CC-BY licence using the Swansea University Research Publications Policy (rights retention). true eng https://creativecommons.org/licenses/by/4.0/deed.en
title Gradients Stand-in for Defending Deep Leakage in Federated Learning
spellingShingle Gradients Stand-in for Defending Deep Leakage in Federated Learning
Freya Hu
Hans Ren
Chen Hu
Yiming Li
Xianghua Xie
title_short Gradients Stand-in for Defending Deep Leakage in Federated Learning
title_full Gradients Stand-in for Defending Deep Leakage in Federated Learning
title_fullStr Gradients Stand-in for Defending Deep Leakage in Federated Learning
title_full_unstemmed Gradients Stand-in for Defending Deep Leakage in Federated Learning
title_sort Gradients Stand-in for Defending Deep Leakage in Federated Learning
author_id_str_mv aa73524c5e3969c88fb7a3a5bde919b1
9e043b899a2b786672a28ed4f864ffcc
55d3ba5f8378c2e3439d7e3962aee726
1b3389c4ef5d90bedb2a23843041ed68
b334d40963c7a2f435f06d2c26c74e11
author_id_fullname_str_mv aa73524c5e3969c88fb7a3a5bde919b1_***_Freya Hu
9e043b899a2b786672a28ed4f864ffcc_***_Hans Ren
55d3ba5f8378c2e3439d7e3962aee726_***_Chen Hu
1b3389c4ef5d90bedb2a23843041ed68_***_Yiming Li
b334d40963c7a2f435f06d2c26c74e11_***_Xianghua Xie
author Freya Hu
Hans Ren
Chen Hu
Yiming Li
Xianghua Xie
author2 Freya Hu
Hans Ren
Chen Hu
Yiming Li
Jingjing Deng
Xianghua Xie
format Conference Paper/Proceeding/Abstract
container_title International Conference on Computing in Natural Sciences, Biomedicine and Engineering
publishDate 2024
institution Swansea University
college_str Faculty of Science and Engineering
hierarchytype
hierarchy_top_id facultyofscienceandengineering
hierarchy_top_title Faculty of Science and Engineering
hierarchy_parent_id facultyofscienceandengineering
hierarchy_parent_title Faculty of Science and Engineering
department_str School of Mathematics and Computer Science - Computer Science{{{_:::_}}}Faculty of Science and Engineering{{{_:::_}}}School of Mathematics and Computer Science - Computer Science
document_store_str 1
active_str 0
description Federated Learning (FL) has become a cornerstone of privacy protection, shifting the paradigm towards localizing sensitive data while only sending model gradients to a central server. This strategy is designed to reinforce privacy protections and minimize the vulnerabilities inherent in centralized data storage systems. Despite its innovative approach, recent empirical studies have highlighted potential weaknesses in FL, notably regarding the exchange of gradients. In response, this studyintroduces a novel, efficacious method aimed at safeguarding against gradient leakage, namely, “AdaDefense”. Following the idea that model convergence can be achieved by using differenttypes of optimization methods, we suggest using a local standin rather than the actual local gradient for global gradient aggregation on the central server. This proposed approach not only effectively prevents gradient leakage, but also ensures that the overall performance of the model remains largely unaffected. Delving into the theoretical dimensions, we explore how gradients may inadvertently leak private information and present a theoretical framework supporting the efficacy of our proposed method. Extensive empirical tests, supported by popular benchmark experiments,validate that our approach maintains model integrity and is robust against gradient leakage, marking an important step in our pursuit of safe and efficient FL.
published_date 2024-09-10T15:53:46Z
_version_ 1814354855547895808
score 11.037603

Similar Items