academic literature review

The research paper discusses the vulnerability of Deep Neural Networks (DNNs) to adversarial examples. Traditional methods assume that attackers can modify any features, neglecting the causal generating process of data,…