一句话：首个针对联邦学习系统特有缺陷的大规模实证研究，基于 395 个真实 bug 归纳症状、易发阶段、根因与修复策略。
联邦学习（FL）允许在分散于各设备的私有数据上训练模型，已被金融、医疗等关键领域采用，但其系统缺陷可能带来经济损失、错误的医疗决策乃至隐私合规风险。现有关于机器学习系统缺陷的研究并未刻画 FL 系统因分布式、隐私保护等特性而产生的独有缺陷。为填补这一空白，作者从 TensorFlow Federated、PySyft、FATE、Flower、PaddleFL 和 Fedlearner 六个主流 FL 框架的 GitHub 与 StackOverflow 中收集了 395 个真实 bug，并人工分析其症状与影响、易发阶段、根因以及修复策略。基于该数据集，论文报告了一系列发现与可操作的启示，用以指导 FL 缺陷的检测与修复实践。
Federated learning (FL) is an emerging machine learning paradigm that aims to address the problem of isolated data islands. To preserve privacy, FL allows machine learning models and deep neural networks to be trained from decentralized data kept privately at individual devices. FL has been increasingly adopted in missioncritical fields such as finance and healthcare. However, bugs in FL systems are inevitable and may result in catastrophic consequences such as financial loss, inappropriate medical decision, and violation of data privacy ordinance. While many recent studies were conducted to understand the bugs in machine learning systems, there is no existing study to characterize the bugs arising from the unique nature of FL systems. To fill the gap, we collected 395 real bugs from six popular FL frameworks (Tensorflow Federated, PySyft, FATE, Flower, PaddleFL, and Fedlearner) in GitHub and StackOverflow, and then manually analyzed their symptoms and impacts, prone stages, root causes, and fix strategies. Furthermore, we report a series of findings and actionable implications that can potentially facilitate the detection of FL bugs.
@inproceedings{DuCC0C023,
title = {Understanding the Bug Characteristics and Fix Strategies of Federated Learning Systems},
author = {Xiaohu Du and Xiao Chen and Jialun Cao and Ming Wen and Shing-Chi Cheung and Hai Jin},
booktitle = {FSE},
year = {2023},
doi = {10.1145/3611643.3616347}
}