Adversarial Model Extraction on Graph Neural Networks

David DeFazio, Arti Ramesh

Along with the advent of deep neural networks came various methods of exploitation, such as fooling the classifier or contaminating its training data. Another such attack is known…