📋 摘要
⭐ 联邦学习样本选择，与SE for AI、测试、公平性等核心方向关联较弱，仅在可信AI层面略擦边。
Sample Selection Using Multi-Task Autoencoders in Federated Learning with Non-IID Data
中文
针对 federated learning 在 non-IID 场景下因冗余、恶意或异常样本导致的模型性能退化与效率低下问题，本文提出了基于 multi-task autoencoder 的样本选择方法用于图像分类任务。该方法通过 loss 与特征分析来估计样本贡献：一方面引入由中心服务器统一管理的无监督离群点检测策略，包括 one-class SVM (OCSVM)、isolation forest (IF) 与 adaptive loss threshold (AT)，用以在客户端过滤噪声样本；另一方面提出由中心服务器控制的 multi-class deep SVDD loss，以增强基于特征的样本选择能力。作者在 CIFAR10 与 MNIST 数据集上，针对不同客户端数量、non-IID 分布及最高 40% 的噪声水平进行了验证。实验结果表明，基于 loss 的样本选择可带来显著的准确率提升，在 CIFAR10 上使用 OCSVM 最高获得 7.02% 的提升，在 MNIST 上使用 AT 最高获得 1.83% 的提升；所提出的 federated SVDD loss 进一步改善了基于特征的样本选择，在 CIFAR10 上结合 OCSVM 可带来最高 0.99% 的准确率提升。结果验证了该方法在不同客户端规模与噪声条件下提升模型准确率的有效性。
English abstract
Federated learning is a machine learning paradigm in which multiple devices collaboratively train a model under the supervision of a central server while ensuring data privacy. However, its performance is often hindered by redundant, malicious, or abnormal samples, leading to model degradation and inefficiency. To overcome these issues, we propose novel sample selection methods for image classification, employing a multitask autoencoder to estimate sample contributions through loss and feature analysis. Our approach incorporates unsupervised outlier detection, using one-class support vector machine (OCSVM), isolation forest (IF), and adaptive loss threshold (AT) methods managed by a central server to filter noisy samples on clients. We also propose a multi-class deep support vector data description (SVDD) loss controlled by a central server to enhance feature-based sample selection. We validate our methods on CIFAR10 and MNIST datasets across varying numbers of clients, non-IID distributions, and noise levels up to 40%. The results show significant accuracy improvements with loss-based sample selection, achieving gains of up to 7.02% on CIFAR10 with OCSVM and 1.83% on MNIST with AT. Additionally, our federated SVDD loss further improves feature-based sample selection, yielding accuracy gains of up to 0.99% on CIFAR10 with OCSVM. These results show the effectiveness of our methods in improving model accuracy across various client counts and noise conditions.
加载中…
点文件 → 加为 tab；按 Esc 关闭