Abstract
背景:In fact, some of the latest findings suggest that the existence of adversarial attacks may be an inherent weakness of deep learning models.
本文: FASER
Github: https://github.com/ise-uiuc/FASER
Task: study the adversarial robustness of neural networks by robust optimization, specify a universal security guarantee that would protect against any adversary, suggest the notion of security against a first-order adversary as a natural and broad security guarantee
- Non-Deterministic Effectiveness Deterministic CDeepFuzz Balancingnon-deterministic effectiveness deterministic non-deterministic effectiveness deterministic balancing deterministic sql function reads re-balancing class-imbalanced classification re-balancing deterministic declaration sql function clientconsumer rebalancing balancing consumer