AI 中文总结
针对联邦学习中基于注意力的自编码器缺乏适配聚合函数的问题，提出两种新型聚合函数，在KDDCUP10数据集上较传统自编码器实现了F1分数最高2.9%、AUC ROC最高5.1%的提升。
AI 中文摘要
在去中心化数据环境中，异常检测对许多机器学习实现而言是一项具有挑战性的任务，尤其是在无法共享数据的场景下。近年来，联邦异常检测取得了进展，部分进展基于自编码器网络的应用。注意力机制的引入提升了自编码器的效率，但由于缺乏适用于这类网络的恰当聚合函数，基于注意力的模型在联邦学习中的应用仍未得到充分发展。本研究提出两种专为基于注意力的自编码器设计的新型聚合函数，能更好地保留这些网络记忆模块中存储的已学习信息。我们在KDDCUP10数据集上对所提方法进行评估，结果显示，与传统自编码器相比，所提方法的F1分数和AUC ROC分别提升了高达2.9%和5.1%。
英文摘要
Outlier detection in decentralized data environments is a challenging task for many machine learning implementations, particularly in settings where data cannot be shared. Recently, there have been advances in federated outlier detection, some of which are based on the use of autoencoder networks. The introduction of attention mechanisms to autoencoders boosts their efficiency. However, the application of attention-based models in federated learning remains underdeveloped due to the absence of proper aggregation functions for these types of networks. In our work, we propose two novel aggregation functions tailored for attention-based autoencoders, which better preserve the learned information stored within the memory modules of these networks. We evaluated our approach on the KDDCUP10 dataset, and we showed that the proposed methods achieve up to 2.9\% and 5.1\% better results for F1 score and AUC ROC respectively when compared to traditional autoencoders.
CommentsAccepted and presented at the 3rd International Conference on Federated Learning Technologies and Applications (FLTA 2025)
Journal refProceedings of the 3rd International Conference on Federated Learning Technologies and Applications (FLTA 2025)
DOI：10.1109/FLTA67013.2025.11336743
