FeDLRT: Federated Dynamical Low-Rank Training
SIAM Journal on Mathematics of Data Science · 31.12.2026 · 8 (4) · 1292-1321
DOI: 10.1137/25m1788142 ↗Kurz gefasst
Fachartikel in der Zeitschrift SIAM Journal on Mathematics of Data Science, online erschienen am 30.10.2026. Verfasst von Steffen Schotthöfer und 1 weiteren Autor:innen. Die Arbeit nennt 63 Literaturangaben. Gefördert von: UT-Battelle, U.S. Department of Energy.
Abstract (Originalsprache)
We propose a federated dynamical low-rank training (FeDLRT) scheme to reduce client compute and communication costs—two significant performance bottlenecks in horizontal federated learning. Our method builds upon dynamical low-rank time integration schemes for manifold-constrained optimization to create a global low-rank basis of network weights, which enables client training on a small coefficient matrix. This global low-rank basis allows us to incorporate a variance correction scheme and prove global loss descent and convergence to a stationary point on a low-rank manifold. FeDLRT features dynamic augmentation and truncation of the low-rank bases to optimize computing and communication resource utilization. Notably FeDLRT only trains a small coefficient matrix per client. We demonstrate the efficiency of FeDLRT in an array of computer vision benchmarks with both independent and identically distributed (IID) and non-IID data and show a reduction of client compute and communication costs by up to an order of magnitude with minimal impacts on global accuracy. FeDLRT performs as well as classical methods such as FedAvg and FedLin, with a fraction of the memory and compute requirements.
Förderung
- UT-Battelle
- U.S. Department of Energy
Zitieren
APA
Schotthöfer, S., & Laiu, M. P. (2026). FeDLRT: Federated Dynamical Low-Rank Training. SIAM Journal on Mathematics of Data Science, 8(4), 1292-1321. https://doi.org/10.1137/25m1788142
Vancouver
Schotthöfer S, Laiu MP. FeDLRT: Federated Dynamical Low-Rank Training. SIAM Journal on Mathematics of Data Science. 2026;8(4):1292-1321. doi:10.1137/25m1788142
BibTeX
@article{Schotthfer2026,
author = {Schotthöfer, Steffen and Laiu, M. Paul},
title = {FeDLRT: Federated Dynamical Low-Rank Training},
journal = {SIAM Journal on Mathematics of Data Science},
year = {2026},
volume = {8},
number = {4},
pages = {1292--1321},
doi = {10.1137/25m1788142}
}
EUVIMED