Adversarial Vulnerability of Neural Networks Increases with Input Dimension
Over the past four years, neural networks have been proven vulnerable to adversarial images: targeted but imperceptible image perturbations lead to drastically different predictions. We show that adversarial vulnerability increases with the gradients of the training objective when viewed as a function of the inputs. For most current network architectures, we prove that the L1-norm of these gradients grows as the square root of the input size. These nets therefore become increasingly vulnerable with growing image size. Our proofs rely on the network’s weight distribution at initialization, but extensive experiments confirm that our conclusions still hold after usual training.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
First-order Adversarial Vulnerability of Neural Networks and Input Dimension
Over the past few years, neural networks were proven vulnerable to adversarial images: targeted but imperceptible image perturbations lead to drastically different predictions. We show that adversarial vulnerability incr…
The Role of Input Dimensionality in the Emergence and Targeted Control of Adversarial Examples
Several theoretical works have tried to explain the adversarial vulnerability of deep neural networks through properties of high-dimensional geometry. However, the assumptions underlying these works are rarely examined e…
On the relationship between class selectivity, dimensionality, and robustness
While the relative trade-offs between sparse and distributed representations in deep neural networks (DNNs) are well-studied, less is known about how these trade-offs apply to representations of semantically-meaningful i…
Defending Against Adversarial Examples by Regularized Deep Embedding
Recent studies have demonstrated the vulnerability of deep convolutional neural networks against adversarial examples. Inspired by the observation that the intrinsic dimension of image data is much smaller than its pixel…
Adversarial AttackAdversarial RobustnessLinking average- and worst-case perturbation robustness via class selectivity and dimensionality
Representational sparsity is known to affect robustness to input perturbations in deep neural networks (DNNs), but less is known about how the semantic content of representations affects robustness. Class selectivity—the…