Backdoor Vulnerabilities in Normally Trained Deep Learning Models. (arXiv:2211.15929v1 [cs.CR])

We conduct a systematic study of backdoor vulnerabilities in normally trained
Deep Learning models. They are as dangerous as backdoors injected by data
poisoning because both can be equally exploited. We leverage 20 different types
of injected backdoor attacks in the literature as the guidance and study their
correspondences in normally trained models, which we call natural backdoor
vulnerabilities. We find that natural backdoors are widely existing, with most
injected backdoor attacks having natural correspondences. We categorize these
natural backdoors …

backdoor deep learning vulnerabilities