Maximization of Average Precision for Deep Learning with Adversarial Ranking Robustness | Read Paper on Bytez