BriefGPT.xyz
Feb, 2025
通过熵正则化解决分布式学习中的标签偏移问题
Addressing Label Shift in Distributed Learning via Entropy Regularization
HTML
PDF
Zhiyuan Wu, Changkyu Choi, Xiangcheng Cao, Volkan Cevher, Ali Ramezani-Kebrya
TL;DR
本研究针对多节点分布式学习中标签偏移问题,提出了一种名为多用途鲁棒标签偏移(VRLS)的方法,旨在通过夏农熵正则化提升训练阶段的标签密度比估计。研究表明,VRLS在处理标签偏移方面显著提升了模型性能,并在多个数据集上超越了基准,展示了其在实际应用中的潜力。
Abstract
We address the challenge of minimizing true risk in multi-node
Distributed Learning
. These systems are frequently exposed to both inter-node and intra-node label shifts, which present a critical obstacle to effectively optimizing
→