A Method to Verify Neural Network Decoders Against Adversarial Attacks | IEEE Journals & Magazine | IEEE Xplore