TY - GEN
T1 - A STOCHASTIC GRADIENT APPROACH FOR COMMUNICATION EFFICIENT CONFEDERATED LEARNING
AU - Wang, Bin
AU - Fang, Jun
AU - Li, Hongbin
AU - Eldar, Yonina C.
N1 - Publisher Copyright:
© 2024 IEEE.
PY - 2024
Y1 - 2024
N2 - In this work, we consider a multi-server federated learning (FL) framework, referred to as Confederated Learning (CFL), in order to accommodate a larger number of users. To reduce the communication overhead of the CFL system, we propose a linearly convergent stochastic gradient method. The proposed algorithm incorporates a conditionally-triggered user selection (CTUS) mechanism as the central component. Simulation results show that it achieves advantageous communication efficiency over GT-SAGA.
AB - In this work, we consider a multi-server federated learning (FL) framework, referred to as Confederated Learning (CFL), in order to accommodate a larger number of users. To reduce the communication overhead of the CFL system, we propose a linearly convergent stochastic gradient method. The proposed algorithm incorporates a conditionally-triggered user selection (CTUS) mechanism as the central component. Simulation results show that it achieves advantageous communication efficiency over GT-SAGA.
KW - communication efficiency
KW - Confederated learning
KW - user selection
UR - http://www.scopus.com/inward/record.url?scp=85195386342&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85195386342&partnerID=8YFLogxK
U2 - 10.1109/ICASSP48485.2024.10448035
DO - 10.1109/ICASSP48485.2024.10448035
M3 - Conference contribution
AN - SCOPUS:85195386342
T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
SP - 5170
EP - 5174
BT - 2024 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2024 - Proceedings
T2 - 49th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2024
Y2 - 14 April 2024 through 19 April 2024
ER -