First, most graph autoencoders ignore the reconstruction of either the graph structure or the node attributes, which often leads to a poor latent representation of the graph-structured data. Second, for existing graph autoencoders models, the encoder and decoder are mainly composed of an initial...
KDD 2022 | Accurate Node Feature Estimation with Structured Variational Graph Autoencoder 文章信息「来源」:Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2022)「标题」:Accurate Node Feature Estimation with Structured Variational Graph Autoencoder「作者」:Yoo Ja...
The graph autoencoder is a type of artificial neural network for unsupervised representation learning on graph-structured data15. The graph autoencoder often has a low-dimensional bottleneck layer so that it can be used as a model for dimensionality reduction. Let the inputs be single-cell graph...
However, there is a lack of theoretical understanding of how masking matters on graph autoencoders (GAEs). In this work, we present masked graph autoencoder (MaskGAE), a self-supervised learning framework for graph-structured data. Different from standard GAEs, MaskGAE adopts masked graph ...
We propose a variational autoencoder model in which both encoder and decoder are graph-structured. Our decoder assumes a sequential ordering of graph extension steps and we discuss and analyze design choices that mitigate the potential downsides of this linearization. Experiments compare our approac...
3) Graph Auto-Encoders GAE 使用GNN结构将网络顶点嵌入到低维向量中。最普遍的解决方案之一是采用多层感知作为输入的编码器[147]。其中,解码器重构顶点的邻域统计。PPMI或第一和第二近邻可以被纳入统计[148], [149]。图表示的深度神经网络(DNGR)采用PPMI。结构性深层网络嵌入(SDNE)采用堆叠式自动编码器来保持一...
Interconnected societies generate large amounts of structured data that frequently stem from observing a common set of objects (or sources) through different modalities. Such multiview datasets are also encountered in many different fields like computational biology [3], acoustics [4], surveillance [5...
We introduce the variational graph auto-encoder (VGAE), a framework for unsupervised learning on graph-structured data based on the variational auto-encoder (VAE). 22 Paper Code Cluster-GCN: An Efficient Algorithm for Training Deep and Large Graph Convolutional Networks ...
We provide an empirical evaluation of our model on five benchmark relational, graph-structured datasets and demonstrate significant improvement over three strong baselines for graph representation learning. Reference code and data are available at https://github.com/vuptran/graph-representation-learning ...
MAERec is a simple yet effective graph masked autoencoder that adaptively and dynamically distills global item transitional information for self-supervised augmentation through a noveladaptive transition path maskingstrategy. It naturally addresses the data scarcity and noise perturbation problems in sequentia...