PuSH - Publikationsserver des Helmholtz Zentrums München

Mächler, L.* ; Ezhov, I.* ; Kofler, F.* ; Shit, S.* ; Paetzold, J.C. ; Loehr, T.* ; Zimmer, C.* ; Wiestler, B.* ; Menze, B.H.*

FedCostWAvg: A new averaging for better federated learning.

Lect. Notes Comput. Sc. 12963 LNCS, 383-391 (2022)
Postprint DOI
Open Access Green
We propose a simple new aggregation strategy for federated learning that won the MICCAI Federated Tumor Segmentation Challenge 2021 (FETS), the first ever challenge on Federated Learning in the Machine Learning community. Our method addresses the problem of how to aggregate multiple models that were trained on different data sets. Conceptually, we propose a new way to choose the weights when averaging the different models, thereby extending the current state of the art (FedAvg). Empirical validation demonstrates that our approach reaches a notable improvement in segmentation performance compared to FedAvg.
Altmetric
Weitere Metriken?
[➜Einloggen]
Zusatzinfos bearbeiten [➜Einloggen]
Publikationstyp Artikel: Journalartikel
Dokumenttyp Wissenschaftlicher Artikel
Korrespondenzautor
Schlagwörter Brain Tumor Segmentation ; Federated Learning ; Machine Learning ; Miccai Challenges ; Mri ; Multi-modal Medical Imaging
ISSN (print) / ISBN 0302-9743
e-ISSN 1611-3349
Zeitschrift Lecture Notes in Computer Science
Quellenangaben Band: 12963 LNCS, Heft: , Seiten: 383-391 Artikelnummer: , Supplement: ,
Verlag Springer
Verlagsort Berlin [u.a.]
Nichtpatentliteratur Publikationen
Institut(e) Institute for Tissue Engineering and Regenerative Medicine (ITERM)