Saliency Methods for Explaining Adversarial Attacks
원문 초록
The classification decisions of neural networks can be misled by small imperceptible perturbations. This work aims to explain the misled classifications using saliency methods. The idea behind saliency methods is to explain the classification decisions of neural networks by creating so-called saliency maps. Unfortunately, a number of recent publications have shown that many of the proposed saliency methods do not provide insightful explanations. A prominent example is Guided Backpropagation (GuidedBP), which simply performs (partial) image recovery. However, our numerical analysis shows the saliency maps created by GuidedBP do indeed contain class-discriminative information. We propose a simple and efficient way to enhance the saliency maps. The proposed enhanced GuidedBP shows the state-of-the-art performance to explain adversary classifications.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
Using Instruction-Tuned Large Language Models to Identify Indicators of Vulnerability in Police Incident Narratives
Harmonic-NAS: Hardware-Aware Multimodal Neural Architecture Search on Resource-constrained Devices
STRAPPER: Preference-based Reinforcement Learning via Self-training Augmentation and Peer Regularization
Unlocking the Use of Raw Multispectral Earth Observation Imagery for Onboard Artificial Intelligence
VNE: An Effective Method for Improving Deep Representation by Manipulating Eigenvalue Distribution
Continuous Indeterminate Probability Neural Network
토론
이 논문과 연결된 포럼 글이 아직 없습니다.