Towards Inductive and Efficient Explanations for Graph Neural Networks | IEEE Journals & Magazine | IEEE Xplore