arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2104.06557cs.LGcs.AIcs.CR
收藏

面向因果联邦学习的增强鲁棒性和隐私性

Towards Causal Federated Learning For Enhanced Robustness and Privacy

MILA(MILA研究所);University of Montreal(蒙特利尔大学)

Sreya Francis, Irene Tenison, Irina Rish

摘要:

联邦学习是一种新兴的隐私保护分布式机器学习方法,通过在参与设备(客户端)上本地执行分布式训练并将本地模型聚合为全局模型来构建共享模型。由于这种方法可以防止数据汇集和汇总,因此有助于在很大程度上降低相关的隐私风险。然而,所有参与客户端的数据样本通常不是独立和相同分布的(非iid),并且学习模型的分布外(OOD)泛化可能很差。除了这个挑战之外,联邦学习还容易受到各种安全攻击,其中一些恶意参与实体致力于插入后门,降低生成的聚合模型以及推断参与实体拥有的数据。在本文中,我们提出了一种学习联邦学习设置中所有参与客户端共有的不变(因果)特征的方法,并实证分析了它如何提高分布(OOD)准确性以及最终学习模型的隐私性。

英文摘要:

Federated Learning is an emerging privacy-preserving distributed machine learning approach to building a shared model by performing distributed training locally on participating devices (clients) and aggregating the local models into a global one. As this approach prevents data collection and aggregation, it helps in reducing associated privacy risks to a great extent. However, the data samples across all participating clients are usually not independent and identically distributed (non-iid), and Out of Distribution(OOD) generalization for the learned models can be poor. Besides this challenge, federated learning also remains vulnerable to various attacks on security wherein a few malicious participating entities work towards inserting backdoors, degrading the generated aggregated model as well as inferring the data owned by participating entities. In this paper, we propose an approach for learning invariant (causal) features common to all participating clients in a federated learning setup and analyze empirically how it enhances the Out of Distribution (OOD) accuracy as well as the privacy of the final learned model.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑