BriefGPT.xyz
Oct, 2021
适应适应性：跨数据源联邦学习个性化学习
Adapt to Adaptation: Learning Personalization for Cross-Silo Federated Learning
HTML
PDF
Jun Luo, Shandong Wu
TL;DR
本文提出了APPLE，这是一个个性化的跨平台联邦学习框架，在不均匀的数据集上实现了最先进的性能表现，并灵活控制了全局和本地目标的焦点。
Abstract
The goal of conventional
federated learning
(FL) is to train a global model for a federation of clients with decentralized data, reducing the systemic privacy risk of centralized training. The distribution shift across
→