DP-KAN：差分隐私科尔莫戈罗夫-阿诺德网络
DP-KAN: Differentially Private Kolmogorov-Arnold Networks
AI总结：
本研究探讨了KAN网络在差分隐私训练中的应用，通过DP-SGD算法实现隐私保护，实验表明其性能与MLP相当，适合差分隐私模型训练。
摘要：
我们研究了科尔莫戈罗夫-阿诺德网络（KAN），该网络最近被提出作为经典多层感知器（MLP）的替代方案，并将其应用于差分隐私模型训练。使用DP-SGD算法，我们证明了KAN可以以直接的方式实现隐私保护，并在多个数据集上评估了其性能。我们的结果表明，KAN的准确性不仅与MLP相当，而且由于隐私约束而经历类似的性能下降，使其适用于差分隐私模型训练。
英文摘要：
We study the Kolmogorov-Arnold Network (KAN), recently proposed as an alternative to the classical Multilayer Perceptron (MLP), in the application for differentially private model training. Using the DP-SGD algorithm, we demonstrate that KAN can be made private in a straightforward manner and evaluated its performance across several datasets. Our results indicate that the accuracy of KAN is not only comparable with MLP but also experiences similar deterioration due to privacy constraints, making it suitable for differentially private model training.
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。
