Adversarial Example Decomposition
Research has shown that widely used deep neural networks are vulnerable to carefully crafted adversarial perturbations. Moreover, these adversarial perturbations often transfer across models. We hypothesize that adversarial weakness is composed of three sources of bias: architecture, dataset, and random initialization. We show that one can decompose adversarial examples into an architecture-dependent component, data-dependent component, and noise-dependent component and that these components behave intuitively. For example, noise-dependent components transfer poorly to all other models, while architecture-dependent components transfer better to retrained models with the same architecture. In addition, we demonstrate that these components can be recombined to improve transferability without sacrificing efficacy on the original model.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Model-Free Adversarial Purification via Coarse-To-Fine Tensor Network Representation
Deep neural networks are known to be vulnerable to well-designed adversarial attacks. Although numerous defense strategies have been proposed, many are tailored to the specific attacks or tasks and often fail to generali…
Adversarial PurificationApplying Tensor Decomposition to image for Robustness against Adversarial Attack
Nowadays the deep learning technology is growing faster and shows dramatic performance in computer vision areas. However, it turns out a deep learning based model is highly vulnerable to some small perturbation called an…
Adversarial AttackDeep LearningTensor DecompositionWaveTransform: Crafting Adversarial Examples via Input Decomposition
Frequency spectrum has played a significant role in learning unique and discriminating features for object recognition. Both low and high frequency information present in images have been extracted and learnt by a host o…
Adversarial DefenseObject RecognitionRepresentation LearningMost Convolutional Networks Suffer from Small Adversarial Perturbations
The existence of adversarial examples is relatively understood for random fully connected neural networks, but much less so for convolutional neural networks (CNNs). The recent work [Daniely, 2025] establishes that adver…
Defending against Patch-Based and Texture-Based Adversarial Attacks with Spectral Decomposition
Adversarial examples present significant challenges to the security of Deep Neural Network (DNN) applications. Specifically, there are patch-based and texture-based attacks that are usually used to craft physical-world a…