arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2402.02230cs.LGcs.AIcs.DC
收藏

联邦学习与差分隐私

Federated Learning with Differential Privacy

EPFL(洛桑联邦理工学院);UCLouvain(鲁汶大学);Universität Zürich(苏黎世大学)

Adrien Banse, Jan Kreischer, Xavier Oliva i Jürgens

AI总结:

本研究通过实证基准测试,探究了联邦学习中客户端数量与差分隐私机制对模型性能的影响,发现非独立同分布和小型数据集在分布式隐私设置下性能下降最大。

摘要:

联邦学习(FL)作为一种分布式机器学习方法,能够显著保护客户端的私有数据,避免其在不同参与方之间共享。然而,通过分析客户端上传的参数权重,私有信息仍可能被泄露。在本报告中,我们展示了关于客户端数量以及添加差分隐私(DP)机制对不同类型数据上模型性能影响的实证基准测试。我们的结果表明,在分布式且差分隐私的设置下,非独立同分布(non-i.i.d)和小型数据集的性能下降最为显著。

英文摘要:

Federated learning (FL), as a type of distributed machine learning, is capable of significantly preserving client's private data from being shared among different parties. Nevertheless, private information can still be divulged by analyzing uploaded parameter weights from clients. In this report, we showcase our empirical benchmark of the effect of the number of clients and the addition of differential privacy (DP) mechanisms on the performance of the model on different types of data. Our results show that non-i.i.d and small datasets have the highest decrease in performance in a distributed and differentially private setting.

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。

↑