BriefGPT.xyz
Dec, 2020
Ditto: 通过个性化功能实现公平且稳健的联邦学习
Federated Multi-Task Learning for Competing Constraints
HTML
PDF
Tian Li, Shengyuan Hu, Ahmad Beirami, Virginia Smith
TL;DR
提出一个称为 Ditto 的基本框架来解决联邦学习中的公平性和鲁棒性竞争的问题，并且在多项实验证明 Ditto 相对于其他竞争方法具有更高的准确性、鲁棒性和公平性。
Abstract
In addition to accuracy,
fairness
and
robustness
are two critical concerns for
federated learning
systems. In this work, we first identify
→