Towards A Critical Evaluation of Robustness for Deep Learning Backdoor Countermeasures

Huming Qiu, Hua Ma, Zhi Zhang +4

Since Deep Learning (DL) backdoor attacks have been revealed as one of the most insidious adversarial attacks, a number of countermeasures have been developed with certain assumpti…