Communication Dans Un Congrès Année : 2025

Trustworthy Efficient Communication for Distributed Learning Using LQ-SGD Algorithm

Résumé

We propose LQ-SGD (Low-Rank Quantized Stochastic Gradient Descent), an efficient communication gradient compression algorithm designed for distributed training. LQSGD further develops on the basis of PowerSGD by incorporating the low-rank approximation and log-quantization techniques, which drastically reduce the communication overhead, while still ensuring the convergence speed of training and model accuracy. In addition, LQ-SGD and other compression-based methods show stronger resistance to gradient inversion than traditional SGD, providing a more robust and efficient optimization path for distributed learning systems.

Dates et versions

hal-05524350 , version 1 (23-02-2026)

Identifiants

Citer

Hongyang Li, Lincen Bai, Caesar Wu, Mohammed Chadli, Saïd Mammar, et al.. Trustworthy Efficient Communication for Distributed Learning Using LQ-SGD Algorithm. 33rd European Signal Processing Conference (EUSIPCO 2025), Sep 2025, Palermo, Italy. pp.1917--1921, ⟨10.23919/EUSIPCO63237.2025.11226134⟩. ⟨hal-05524350⟩
36 Consultations
0 Téléchargements

Altmetric

Partager

  • More