… APN, GAP simply uses graph neighborhood of nodes. That is, for every node in the graph
we define a graph neighborhood function to build a fixed size neighborhood sequence, which …

… attentive pooling graph network (HAPGN) which utilizes the gated graph attention network
(GGAN) and hierarchical graph … where Ni is the neighborhood (256-nearest neighbor nodes in …

… graph pooling module can be interleaved with the graph convolution layer to coarsen the …
A semi-supervised graph attentive network for financial fraud detection. In 2019 IEEE …

… In this paper, we propose a multi-modal attentive graph pooling … answer into a multi-modal
content graph, which can preserve the … content and model them into the multi-modal graph. An …

… implicit correlations of neighbors in graph embedding … neighbors, we further stack more
neighbor-aware graph attention layer to gather the influential signals from higher-order neighbors…

… graph pooling layer. To reduce the impact of spurious edges on graph clustering, we design
a novel neighbor cluster pooling … of nodes and their nearest neighbors to the cluster centers. …

… We define the immediate (or first order) neighborhood of a node vi as Ni = 1vj|(vi,vj) ∈ El. …
input graph G to be undirected for the rest of the paper, but extending it for directed graphs is …

… node classification, rather than text classification; (2) We use one-dimensional convolution
and pooling layers to extract node features, instead of looking up the embeddings from some …

… graph, analogous to the neighborhood aggregator in graph convolution. Therefore, a good
graph pooling method should encourage graphs … the self-attentive mechanism to learn node …

… neighborhood structure, we propose a novel Graph embedding model with Hierarchical
Attentive … However, such graph pooling methods lack necessary control on the weight matrix to …