Malmö University Publications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Federated Learning and Privacy, Challenges, Threat and Attack Models, and Analysis
Department of Computer Science and Engineering, Indian Institute of Information Technology, Kerala, Kottayam, India.
Department of Computer Science and Engineering, Indian Institute of Information Technology, Kerala, Kottayam, India.
Malmö University, Faculty of Technology and Society (TS), Department of Computer Science and Media Technology (DVMT). Malmö University, Internet of Things and People (IOTAP).ORCID iD: 0000-0002-2763-8085
2024 (English)In: Federated Learning: Principles, Paradigms, and Applications / [ed] Jayakrushna Sahoo; Mariya Ouaissa; Akarsh K. Nair, CRC Press, 2024, p. 183-212Chapter in book (Refereed)
Abstract [en]

The advent of intertwined technology, conjoined with powerful centralized machine algorithms, spawns the need for privacy. The efficiency and accuracy of any Machine Learning (ML) algorithm are proportional to the quantity and quality of data collected for training, which could often compromise the data subject’s privacy. Federated Learning (FL) or collaborative learning is a branch of Artificial Intelligence (AI) that decentralizes ML algorithms across edge devices or local servers. This chapter discusses privacy threat models in ML and expounds on FL as a Privacy-preserving Machine Learning (PPML) system by distinguishing FL from other decentralized ML algorithms. We elucidate the comprehensive secure FL framework with Horizontal FL, Vertical FL, and Federated Transfer Learning that mitigates privacy issues. For privacy preservation, FL extends its capacity to incorporate Differential Privacy (DP) techniques to provide quantifiable measures on data anonymization. We have also discussed the concepts in FL that comprehend Local Differential Privacy (LDP) and Global Differential Privacy (GDP). The chapter concludes with 184open research problems and challenges of FL as PPML with implications, limitations, and future scope. 

Place, publisher, year, edition, pages
CRC Press, 2024. p. 183-212
National Category
Computer Sciences
Identifiers
URN: urn:nbn:se:mau:diva-70152DOI: 10.1201/9781003497196-8Scopus ID: 2-s2.0-85199107660ISBN: 9781774916384 (print)ISBN: 9781003497196 (print)OAI: oai:DiVA.org:mau-70152DiVA, id: diva2:1888231
Available from: 2024-08-12 Created: 2024-08-12 Last updated: 2024-08-12Bibliographically approved

Open Access in DiVA

No full text in DiVA

Other links

Publisher's full textScopus

Authority records

Malekian, Reza

Search in DiVA

By author/editor
Malekian, Reza
By organisation
Department of Computer Science and Media Technology (DVMT)Internet of Things and People (IOTAP)
Computer Sciences

Search outside of DiVA

GoogleGoogle Scholar

doi
isbn
urn-nbn

Altmetric score

doi
isbn
urn-nbn
Total: 52 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf