Zur Hauptnavigation wechseln Zur Suche wechseln Zum Hauptinhalt wechseln

Poster Abstract: OrderlessFL: A CRDT-Enabled Permissioned Blockchain for Federated Learning

  • Technische Universität München
  • University of Toronto

Publikation: Beitrag in Buch/Bericht/KonferenzbandKonferenzbeitragBegutachtung

1 Zitat (Scopus)

Abstract

Industries produce a large amount of data that can improve Machine Learning models. However, due to privacy issues, the data cannot be shared. Several Federated Learning (FL) systems have been introduced as private alternatives without considering Byzantine actors. Also, these systems are affected by the gradient staleness problem. Several blockchain-based FL systems are introduced to address Byzantine actors, which rely on Proof-of-Work-based (PoW) protocols and suffer from their limitations. We introduce OrderlessFL, a safe permissioned blockchain-based FL system using flCRDT, a CRDT for concurrent ML training and mitigating gradient staleness.

OriginalspracheEnglisch
TitelMiddleware 2022 - Proceedings of the 23rd International Middleware Conference Demos and Posters, Part of Middleware 2022
Herausgeber (Verlag)Association for Computing Machinery, Inc
Seiten7-8
Seitenumfang2
ISBN (elektronisch)9781450399319
DOIs
PublikationsstatusVeröffentlicht - 7 Nov. 2022
Veranstaltung23rd International Middleware Conference, Middleware 2022 - Part of Middleware 2022 - Quebec, Kanada
Dauer: 7 Nov. 202211 Nov. 2022

Publikationsreihe

NameMiddleware 2022 - Proceedings of the 23rd International Middleware Conference Demos and Posters, Part of Middleware 2022

Konferenz

Konferenz23rd International Middleware Conference, Middleware 2022 - Part of Middleware 2022
Land/GebietKanada
OrtQuebec
Zeitraum7/11/2211/11/22

Fingerprint

Untersuchen Sie die Forschungsthemen von „Poster Abstract: OrderlessFL: A CRDT-Enabled Permissioned Blockchain for Federated Learning“. Zusammen bilden sie einen einzigartigen Fingerprint.

Dieses zitieren