Scalable attack on graph data by injecting vicious nodes
Sign Up to like & getrecommendations! Published in 2020 at "Data Mining and Knowledge Discovery"
DOI: 10.1007/s10618-020-00696-7
Abstract: Recent studies have shown that graph convolution networks (GCNs) are vulnerable to carefully designed attacks, which aim to cause misclassification of a specific node on the graph with unnoticeable perturbations. However, a vast majority of… read more here.
Keywords: attack; injecting vicious; attack graph; scalable attack ... See more keywords