arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2401.02453cs.CRcs.DCcs.LG
收藏

联邦学习中的自适应差分隐私:一种基于优先级的方法

Adaptive Differential Privacy in Federated Learning: A Priority-Based Approach

Isfahan University of Technology(伊斯法罕理工大学)

Mahtab Talaei, Iman Izadi

AI总结:

针对联邦学习中差分隐私噪声注入导致模型精度下降的问题,提出基于特征相对重要性的自适应噪声添加方法,通过向重要参数加较少噪声、不重要参数加较多噪声,在保护隐私的同时有效维持模型精度。

摘要:

联邦学习(FL)作为分布式机器学习(ML)的新兴分支之一,通过无需直接访问本地数据集的隐私保护过程来开发全局模型。然而,对客户端和服务器之间传输的模型更新(例如深度神经网络中的梯度更新)的访问,可能会向攻击者泄露敏感信息。差分隐私(DP)提供了一个框架,通过向参数添加一定量的噪声来提供隐私保证。这种方法虽然在隐私方面很有效,但由于引入噪声,会对模型性能产生不利影响。因此,始终需要在噪声注入和牺牲的准确性之间找到平衡。为了应对这一挑战,我们提出了FL中的自适应噪声添加方法,该方法根据特征的相对重要性来决定注入噪声的值。在这里,我们首先提出了两种对深度神经网络模型中的特征进行优先级排序的有效方法,然后基于此信息扰动模型的权重。具体而言,我们试图弄清楚“向不重要的参数添加更多噪声,向重要的参数添加更少噪声”的想法,能否在保护隐私的同时有效挽救模型准确性。我们的实验在某些条件下证实了这一说法。注入的噪声量、涉及的参数比例以及全局迭代次数可以显著改变输出。通过考虑数据集属性来仔细选择参数,可以在没有严重精度损失的情况下提高隐私性,而糟糕的选择会使模型性能变差。

英文摘要:

Federated learning (FL) as one of the novel branches of distributed machine learning (ML), develops global models through a private procedure without direct access to local datasets. However, access to model updates (e.g. gradient updates in deep neural networks) transferred between clients and servers can reveal sensitive information to adversaries. Differential privacy (DP) offers a framework that gives a privacy guarantee by adding certain amounts of noise to parameters. This approach, although being effective in terms of privacy, adversely affects model performance due to noise involvement. Hence, it is always needed to find a balance between noise injection and the sacrificed accuracy. To address this challenge, we propose adaptive noise addition in FL which decides the value of injected noise based on features' relative importance. Here, we first propose two effective methods for prioritizing features in deep neural network models and then perturb models' weights based on this information. Specifically, we try to figure out whether the idea of adding more noise to less important parameters and less noise to more important parameters can effectively save the model accuracy while preserving privacy. Our experiments confirm this statement under some conditions. The amount of noise injected, the proportion of parameters involved, and the number of global iterations can significantly change the output. While a careful choice of parameters by considering the properties of datasets can improve privacy without intense loss of accuracy, a bad choice can make the model performance worse.

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。

↑