MISA: Online Defense of Trojaned Models using Misattributions
원문 초록
Recent studies have shown that neural networks are vulnerable to Trojan attacks, where a network is trained to respond to specially crafted trigger patterns in the inputs in specific and potentially malicious ways. This paper proposes MISA, a new online approach to detect Trojan triggers for neural networks at inference time. Our approach is based on a novel notion called misattributions, which captures the anomalous manifestation of a Trojan activation in the feature space. Given an input image and the corresponding output prediction, our algorithm first computes the model's attribution on different features. It then statistically analyzes these attributions to ascertain the presence of a Trojan trigger. Across a set of benchmarks, we show that our method can effectively detect Trojan triggers for a wide variety of trigger patterns, including several recent ones for which there are no known defenses. Our method achieves 96% AUC for detecting images that include a Trojan trigger without any assumptions on the trigger pattern.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
Evaluating Adversarial Attacks on Traffic Sign Classifiers beyond Standard Baselines
CausalDiff: Causality-Inspired Disentanglement via Diffusion Model for Adversarial Defense
TSCLIP: Robust CLIP Fine-Tuning for Worldwide Cross-Regional Traffic Sign Recognition
Watch Out! Simple Horizontal Class Backdoor Can Trivially Evade Defense
Benchmarking Local Robustness of High-Accuracy Binary Neural Networks for Enhanced Traffic Sign Recognition
Adversarial Robustness Certification for Bayesian Neural Networks
토론
이 논문과 연결된 포럼 글이 아직 없습니다.