4 Federated Personalized Subgraph Learning 4.1 Subgraph Similarity Estimation Model Parameters for Subgraph Similarities(参数相似度) 首先考虑使用模型参数之间的相似度进行度量,即S=\frac{\theta_i \cdot \theta_j}{\Vert \theta_i\Vert
论文链接:Personalized Subgraph Federated Learning (mlr.press) 代码链接:GitHub - JinheonBaek/FED-PUB: Official Code Repository for the paper - Personalized Subgraph Federated Learning (ICML 2023) 1、引言 大多数现有的图神经网络(GNNs)都着重于单个图,其节点和边从多个来源收集,并存储在中央服务器中。然而...
For each subgraph, the violin plot display the distribution of daily average power usage, with the median (50th percentile) indicated by a solid line, and the interquartile range (25th and 75th percentiles) visualized within the plot. Table 1. Data description. #HouseholdAppliances Empty Cell...