Explaining What a Neural Network has Learned: Toward Transparent Classification | IEEE Conference Publication | IEEE Xplore