Watermarking Graph Neural Networks by Random Graphs

Many learning tasks require us to deal with graph data which contains rich relational information among elements, leading increasing graph neural network (GNN) models to be deployed in industrial products for improving service quality. However, they also raise challenges to model authentication. It is necessary to protect the ownership of the GNN models, which motivates us to watermark GNN models. In this work, an Erdos-Renyi (ER) random graph with random node feature vectors and labels is randomly generated as a trigger to train the GNN to be protected together with the normal samples. During model training, the secret watermark is embedded into the label predictions of graph nodes. During model verification, by activating a marked GNN with the trigger ER graph, the watermark can be reconstructed from the output to verify the ownership. Since the ER graph was randomly generated, by feeding it to a non-marked GNN, the label predictions of graph nodes are random, resulting in a low false alarm rate (of proposed work). Experimental results have also shown that, the performance of a marked GNN on its original task will not be impaired. And, it is robust against model compression and fine-tuning, which has shown superiority and applicability.