Defenses against adversarial examples, such as adversarial training, are typically tailored to a single perturbation type (e.g., small Linf-noise). For other perturbations, these defenses offer no guarantees and, at times, even increase the model’s vulnerability. Our aim is to understand the reasons underlying this robustness trade-off, and to train models that are simultaneously robust to multiple perturbation types. We prove that a trade-off in robustness to different types of Lp-bounded and spatial perturbations must exist in a natural and simple statistical setting. We corroborate our formal analysis by demonstrating similar robustness trade-offs on MNIST and CIFAR10. Building upon new multi-perturbation adversarial training schemes, and a novel efficient attack for finding L1-bounded adversarial examples, we show that no model trained against multiple attacks achieves robustness competitive with that of models trained on each attack individually. In particular, we uncover a pernicious gradient-masking phenomenon on MNIST, which causes adversarial training with first-order Linf, L1 and L2 adversaries to achieve merely 50% accuracy. Our results question the viability and computational scalability of extending adversarial robustness, and adversarial training, to multiple perturbation types.
more »
« less
Dual Manifold Adversarial Robustness: Defense against Lp and non-Lp Adversarial Attacks
Adversarial training is a popular defense strategy against attack threat models with bounded Lp norms. However, it often degrades the model performance on normal images and the defense does not generalize well to novel attacks. Given the success of deep generative models such as GANs and VAEs in characterizing the underlying manifold of images, we investigate whether or not the aforementioned problems can be remedied by exploiting the underlying manifold information. To this end, we construct an "On-Manifold ImageNet" (OM-ImageNet) dataset by projecting the ImageNet samples onto the manifold learned by StyleGSN. For this dataset, the underlying manifold information is exact. Using OM-ImageNet, we first show that adversarial training in the latent space of images improves both standard accuracy and robustness to on-manifold attacks. However, since no out-of-manifold perturbations are realized, the defense can be broken by Lp adversarial attacks. We further propose Dual Manifold Adversarial Training (DMAT) where adversarial perturbations in both latent and image spaces are used in robustifying the model. Our DMAT improves performance on normal images, and achieves comparable robustness to the standard adversarial training against Lp attacks. In addition, we observe that models defended by DMAT achieve improved robustness against novel attacks which manipulate images by global color shifts or various types of image filtering. Interestingly, similar improvements are also achieved when the defended models are tested on out-of-manifold natural images. These results demonstrate the potential benefits of using manifold information in enhancing robustness of deep learning models against various types of novel adversarial attacks.
more »
« less
- Award ID(s):
- 1942230
- NSF-PAR ID:
- 10207644
- Date Published:
- Journal Name:
- Advances in Neural Information Processing Systems Foundation (NeurIPS)
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Modern image classification systems are often built on deep neural networks, which suffer from adversarial examples—images with deliberately crafted, imperceptible noise to mislead the network’s classification. To defend against adversarial examples, a plausible idea is to obfuscate the network’s gradient with respect to the input image. This general idea has inspired a long line of defense methods. Yet, almost all of them have proven vulnerable. We revisit this seemingly flawed idea from a radically different perspective. We embrace the omnipresence of adversarial examples and the numerical procedure of crafting them, and turn this harmful attacking process into a useful defense mechanism. Our defense method is conceptually simple: before feeding an input image for classification, transform it by finding an adversarial example on a pre- trained external model. We evaluate our method against a wide range of possible attacks. On both CIFAR-10 and Tiny ImageNet datasets, our method is significantly more robust than state-of-the-art methods. Particularly, in comparison to adversarial training, our method offers lower training cost as well as stronger robustness.more » « less
-
As machine learning (ML) systems become pervasive, safeguarding their security is critical. However, recently it has been demonstrated that motivated adversaries are able to mislead ML systems by perturbing test data using semantic transformations. While there exists a rich body of research providing provable robustness guarantees for ML models against ℓp norm bounded adversarial perturbations, guarantees against semantic perturbations remain largely underexplored. In this paper, we provide TSS -- a unified framework for certifying ML robustness against general adversarial semantic transformations. First, depending on the properties of each transformation, we divide common transformations into two categories, namely resolvable (e.g., Gaussian blur) and differentially resolvable (e.g., rotation) transformations. For the former, we propose transformation-specific randomized smoothing strategies and obtain strong robustness certification. The latter category covers transformations that involve interpolation errors, and we propose a novel approach based on stratified sampling to certify the robustness. Our framework TSS leverages these certification strategies and combines with consistency-enhanced training to provide rigorous certification of robustness. We conduct extensive experiments on over ten types of challenging semantic transformations and show that TSS significantly outperforms the state of the art. Moreover, to the best of our knowledge, TSS is the first approach that achieves nontrivial certified robustness on the large-scale ImageNet dataset. For instance, our framework achieves 30.4% certified robust accuracy against rotation attack (within ±30∘) on ImageNet. Moreover, to consider a broader range of transformations, we show TSS is also robust against adaptive attacks and unforeseen image corruptions such as CIFAR-10-C and ImageNet-C.more » « less
-
Deep learning models are vulnerable to adversarial examples. Most of current adversarial attacks add pixel-wise perturbations restricted to some L^p-norm, and defense models are evaluated also on adversarial examples restricted inside L^p-norm balls. However, we wish to explore adversarial examples exist beyond L^p-norm balls and their implications for attacks and defenses. In this paper, we focus on adversarial images generated by transformations. We start with color transformation and propose two gradient-based attacks. Since L^p-norm is inappropriate for measuring image quality in the transformation space, we use the similarity between transformations and the Structural Similarity Index. Next, we explore a larger transformation space consisting of combinations of color and affine transformations. We evaluate our transformation attacks on three data sets --- CIFAR10, SVHN, and ImageNet --- and their corresponding models. Finally, we perform retraining defenses to evaluate the strength of our attacks. The results show that transformation attacks are powerful. They find high-quality adversarial images that have higher transferability and misclassification rates than C&W's L^p attacks, especially at high confidence levels. They are also significantly harder to defend against by retraining than C&W's L^p attacks. More importantly, exploring different attack spaces makes it more challenging to train a universally robust model.more » « less
-
null (Ed.)In the last couple of years, several adversarial attack methods based on different threat models have been proposed for the image classification problem. Most existing defenses consider additive threat models in which sample perturbations have bounded L_p norms. These defenses, however, can be vulnerable against adversarial attacks under non-additive threat models. An example of an attack method based on a non-additive threat model is the Wasserstein adversarial attack proposed by Wong et al. (2019), where the distance between an image and its adversarial example is determined by the Wasserstein metric ("earth-mover distance") between their normalized pixel intensities. Until now, there has been no certifiable defense against this type of attack. In this work, we propose the first defense with certified robustness against Wasserstein Adversarial attacks using randomized smoothing. We develop this certificate by considering the space of possible flows between images, and representing this space such that Wasserstein distance between images is upper-bounded by L_1 distance in this flow-space. We can then apply existing randomized smoothing certificates for the L_1 metric. In MNIST and CIFAR-10 datasets, we find that our proposed defense is also practically effective, demonstrating significantly improved accuracy under Wasserstein adversarial attack compared to unprotected models.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- The DOI is not currently available.
