Improving adversarial robustness using knowledge distillation guided by attention information bottleneck

Deep neural networks have shown outstanding performance in various areas, but adversarial examples can easily fool them. Although strong adversarial attacks have defeated diverse adversarial defense methods, adversarial training, which augments training data with adversarial examples, remains an effective defense strategy. To further improve adversarial robustness, this paper exploits adversarial examples of adversarial examples. We observe that these doubly adversarial examples tend to return to the original prediction on the clean images but sometimes drift toward other classes. From this finding, we propose a regularization loss that prevents these drifts, which mitigates the vulnerability against multi-targeted attacks. Experimental results on the CIFAR-10 and CIFAR-100 datasets empirically show that the proposed loss improves adversarial robustness.

Adversarial training (AT) is widely considered as the most promising strategy to defend against adversarial attacks and has drawn increasing interest from researchers. However, the existing AT methods still suffer from two challenges. First, they are unable to handle unrestricted adversarial examples (UAEs), which are built from scratch, as opposed to restricted adversarial examples (RAEs), which are created by adding perturbations bound by an lp norm to observed examples. Second, the existing AT methods often achieve adversarial robustness at the expense of standard generalizability (i.e., the accuracy on natural examples) because they make a tradeoff between them. To overcome these challenges, we propose a unique viewpoint that understands UAEs as imperceptibly perturbed unobserved examples. Also, we find that the tradeoff results from the separation of the distributions of adversarial examples and natural examples. Based on these ideas, we propose a novel AT approach called Provable Unrestricted Adversarial Training (PUAT), which can provide a target classifier with comprehensive adversarial robustness against both UAE and RAE, and simultaneously improve its standard generalizability. Particularly, PUAT utilizes partially labeled data to achieve effective UAE generation by accurately capturing the natural data distribution through a novel augmented triple-GAN. At the same time, PUAT extends the traditional AT by introducing the supervised loss of the target classifier into the adversarial loss and achieves the alignment between the UAE distribution, the natural data distribution, and the distribution learned by the classifier, with the collaboration of the augmented triple-GAN. Finally, the solid theoretical analysis and extensive experiments conducted on widely-used benchmarks demonstrate the superiority of PUAT.

AMAKD: Adversarial multi-modal attention-based knowledge distillation for robust behaviour anomaly detection in real-world environments

Deep neural networks (DNNs) have gained widespread adoption in computer vision. Unfortunately, state-of-the-art DNNs are vulnerable to adversarial example (AE) attacks, where an adversary introduces imperceptible perturbations to a test example for defrauding DNNs. The obstacles have urged intensive research on improving the DNN robustness via adversarial training, that is, the clean data set is blended with adversarial examples to carry out training. However, the adversarial example attack technologies are open-ended, and the adversarial training is insufficient to focus on improving robustness performance. To circumvent this limitation, we mitigate adversarial example attacks from another perspective, which aims at detecting adversarial examples. Feature autoencoder detector (FADetector), a novel defense framework that exploits feature knowledge is proposed. One of the hallmarks of FADetector is to not involve adversarial examples to train the detector. Our extensive evaluation on MNIST and CIFAR-10 data sets demonstrates that our defense outperforms the conventional autoencoder detectors in terms of detection accuracy.

Adversarial examples are crucial tools for assessing the robustness of deep neural networks (DNNs) and revealing potential security vulnerabilities. Adversarial example generation methods based on Generative Adversarial Networks (GANs) have made significant progress in generating image adversarial examples, but still suffer from insufficient sparsity and transferability. To address these issues, this study proposes a novel semi-white-box untargeted adversarial example generation method named Wavelet-AdvGAN, with an explicit threat model defined as follows. Specifically, the attack is strictly untargeted without predefined target categories, aiming solely to mislead DNNs into classifying adversarial examples into any category other than the original label. It adopts a semi-white-box setting where attackers are denied access to the target model’s private information. Regarding the generator’s information dependence, the training phase only utilizes public resources (i.e., the target model’s public architecture and CIFAR-10 public training data), while the test phase generates adversarial examples through one-step feedforward of clean images without interacting with the target model. The method incorporates a Frequency Sub-band Difference (FSD) module and a Wavelet Transform Local Feature (WTLF) extraction module, evaluating the differences between original and adversarial examples from the frequency domain perspective. This approach constrains the magnitude of perturbations, reinforces feature regions, and further enhances the attack effectiveness, thereby improving the sparsity and transferability of adversarial examples. Experimental results demonstrate that the Wavelet-AdvGAN method achieves an average increase of 1.26% in attack success rates under two defense strategies—data augmentation and adversarial training. Additionally, the adversarial transferability improves by an average of 2.7%. Moreover, the proposed method exhibits a lower l0 norm, indicating better perturbation sparsity. Consequently, it effectively evaluates the robustness of deep neural networks.

State-of-the-art Deep Neural Networks (DNNs) are promoting medical image processing. However, DNNs are susceptible to adversarial attacks, which could significantly deteriorate model performances and pose a threat to clinical diagnoses. One common method for prevention is through adversarial training, which is highly dependent on harnessing adversarial examples during the training stage. However, adversarial medical examples generated by many existing works are too perceptible to be adversarial examples. To improve the imperceptibility, we proposed a Regional Saliency Map Attack that generates an adversarial example by only perturbing a small number of pixels. Extensive experiments have shown that, on average, our method caused the same degradation in model performance by quantitatively less perceptible perturbations. Visualisations have also verified that the improvement in imperceptibility in an image is both global and regional.

Adversarial examples have raised public concern about the robustness of deep neural networks (DNNs). One universal approach to enhance the robustness is adversarial training which essentially augments the training data. However, adversarial training succeeded only to a very limited extent. This limited progress is partly due to the lack of interpretation and understanding of the robustness of DNNs. In this context, we try to explain the adversarial robustness by embedding the sample points onto a hypersphere, which naturally provides an interpretable metric for the distance between sample points. Different from the empirical intuition, it is observed that adversarial trained models show complex patterns when facing different datasets and training configurations. Observations and explanations about robustness and model behavior are made from the aspect of the distances between samples points. Lastly, we discuss the degradation of standard accuracy in the adversarial trained models and provide possible solutions.

With the widespread applications of Deep Neural Networks (DNNs), the safety of DNNs has become a significant issue. The vulnerability of the neural networks against adversarial examples deepens concerns about the safety of DNNs applications. This paper proposed a novel defence method to improve the adversarial robustness of DNN classifiers without using adversarial training. This method introduces two new loss functions. First, a zero-cross-entropy loss is used to punish overconfidence and find the appropriate confidence for different instances. Second, a logit balancing loss is proposed to protect DNNs from non-targeted attacks by regularising incorrect classes’ logits distribution. This method achieved competitive adversarial robustness compared to advanced adversarial training methods. Meanwhile, a novel robustness diagram is proposed to analyse, interpret and visualise the robustness of DNN classifiers against adversarial attacks. Furthermore, a Log-Softmax-pattern-based adversarial attack detection method is proposed. This detection method can distinguish clean inputs and multiple adversarial attacks via one multi-classification MLP. In particular, it is state-of-the-art in identifying white-box gradient-based attacks; it achieved at least 95.5% accuracy for classifying four white-box gradient-based attacks with maximum 0.1% false positive ratio.

GradFuzz: Fuzzing deep neural networks with gradient vector coverage for adversarial examples

Adversarial examples cause neural networks to produce incorrect outputs with high confidence. Although adversarial training is one of the most effective forms of defense against adversarial examples, unfortunately, a large gap exists between test accuracy and training accuracy in adversarial training. In this paper, we identify Adversarial Feature Overfitting (AFO), which may cause poor adversarially robust generalization, and we show that adversarial training can overshoot the optimal point in terms of robust generalization, leading to AFO in our simple Gaussian model. Considering these theoretical results, we present soft labeling as a solution to the AFO problem. Furthermore, we propose Adversarial Vertex mixup (AVmixup), a soft-labeled data augmentation approach for improving adversarially robust generalization. We complement our theoretical analysis with experiments on CIFAR10, CIFAR100, SVHN, and Tiny ImageNet, and show that AVmixup significantly improves the robust generalization performance and that it reduces the trade-off between standard accuracy and adversarial robustness.

This paper has four main goals. First, we show how we solved the CHES 2018 AES challenge in the contest using essentially a linear classifier combined with a SAT solver and a custom error correction method. This part of the paper has previously appeared in a preprint by the current authors (e-print report 2019/094) and later as a contribution to a preprint write-up of the solutions by the winning teams (e-print report 2019/860).Second, we develop a novel deep neural network architecture for side-channel analysis that completely breaks the AES challenge, allowing for fairly reliable key recovery with just a single trace on the unknown-device part of the CHES challenge (with an expected success rate of roughly 70% if about 100 CPU hours are allowed for the equation solving stage of the attack). This solution significantly improves upon all previously published solutions of the AES challenge, including our baseline linear solution.Third, we consider the question of leakage attribution for both the classifier we used in the challenge and for our deep neural network. Direct inspection of the weight vector of our machine learning model yields a lot of information on the implementation for our linear classifier. For the deep neural network, we test three other strategies (occlusion of traces; inspection of adversarial changes; knowledge distillation) and find that these can yield information on the leakage essentially equivalent to that gained by inspecting the weights of the simpler model.Fourth, we study the properties of adversarially generated side-channel traces for our model. Partly reproducing recent computer vision work by Ilyas et al. in our application domain, we find that a linear classifier that generalizes to an unseen device much better than our linear baseline can be trained using only adversarial examples (fresh random keys, adversarially perturbed traces) for our deep neural network. This gives a new way of extracting human-usable knowledge from a deep side channel model while also yielding insights on adversarial examples in an application domain where relatively few sources of spurious correlations between data and labels exist.The experiments described in this paper can be reproduced using code available at https://github.com/agohr/ches2018.KeywordsPower analysisMachine learningDeep learningSAT solver

Despite deep neural networks (DNNs) have been widely applied in remote sensing (RS) scene classification and achieved satisfying performance, the vulnerability of DNNs towards adversarial examples significantly degrades their performance. Moreover, the relatively limited labeled samples of RS scene classification make DNNs more likely to overfit, leading to weak generalizability and noise sensitivity. This may result in DNNs being more vulnerable to adversarial examples. Consequently, the defense of adversarial examples is of crucial importance to improve both the generalizability and robustness of DNNs in the RS scene classification task. However, few studies have been conducted on defense for RS scene classification, especially ignoring the intrinsic characteristics of RS images. In this paper, an effective defense framework for RS scene classification, named reconstruction-assisted and distance-optimized adversarial training (RDAT), is proposed to defend adversarial examples. In order to solve the problems caused by high interclass similarity, a distance-optimized (DO) strategy is designed for adversarial training to strengthen the learning of underfitting content, increase the interclass distance, and improve the robustness of the networks. Furthermore, in order to generate high quality samples for adversarial training, a reconstruction-assisted (RA) block is proposed to eliminate adversarial perturbations in adversarial examples. Specifically, in this block, by swin transformer (SwinT) block and multi-scale convolution (MSC) block, SwinT-MSC-UNet (SMUNet) is constructed to fully extract global and multi-scale local features to adapt to the characteristics of RS images with large variance of ground object scales. Extensive experiments on the benchmark datasets, i.e., UC Merced (UCM) and Aerial Image Dataset (AID), have demonstrate that the proposed RDAT can effectively resist multiple adversarial attacks and yield superior results than other defense methods for RS scene classification.

State-of-the-art neural network models are actively used in various fields, but it is well-known that they are vulnerable to adversarial example attacks. Throughout the efforts to make the models robust against adversarial example attacks, it has been found to be a very difficult task. While many defense approaches were shown to be not effective, adversarial training remains as one of the promising methods. In adversarial training, the training data are augmented by “adversarial” samples generated using an attack algorithm. If the attacker uses a similar attack algorithm to generate adversarial examples, the adversarially trained network can be quite robust to the attack. However, there are numerous ways of creating adversarial examples, and the defender does not know what algorithm the attacker may use. A natural question is: Can we use adversarial training to train a model robust to multiple types of attack? Previous work have shown that, when a network is trained with adversarial examples generated from multiple attack methods, the network is still vulnerable to white-box attacks where the attacker has complete access to the model parameters. In this paper, we study this question in the context of black-box attacks, which can be a more realistic assumption for practical applications. Experiments with the MNIST dataset show that adversarially training a network with an attack method helps defending against that particular attack method, but has limited effect for other attack methods. In addition, even if the defender trains a network with multiple types of adversarial examples and the attacker attacks with one of the methods, the network could lose accuracy to the attack if the attacker uses a different data augmentation strategy on the target network. These results show that it is very difficult to make a robust network using adversarial training, even for black-box settings where the attacker has restricted information on the target network.

We study robustness issues around DNN deployments on resource constrained systems, under both environmental and adversarial input adaptation. We propose a means of compressing a Radio Frequency (RF) deep neural network architecture through weight pruning, and provide a systems-level analysis of the implementation of such a pruned architecture at resource-constrained edge devices. In particular, we jointly train and sparsify neural networks tailored to edge hardware implementations. Under only negligible accuracy loss (less than 1%), we can achieve at most 27.2x pruning rate for 50-device classification. We demonstrate the efficacy of our approach over multiple edge hardware platforms: our method yields significant inference speedups, 11.5x on an FPGA and 3x on a smartphone, as well as high efficiency. Next, we propose a new learn-prune-share (LPS) algorithm for achieving robustness to environment adaptation in the field of lifelong learning. Our method maintains a parsimonious neural network model and achieves exact no forgetting by splitting the network into task-specific partitions via a weight pruning strategy optimized by the Alternating Direction Methods of Multipliers (ADMM). Moreover, a novel selective knowledge sharing scheme is integrated seamlessly into the ADMM optimization framework to address knowledge reuse. We show that our approach achieves significant improvement over the state-of-the-art methods on multiple real-life datasets. Furthermore, we investigate the Hilbert-Schmidt Information Bottleneck as regularizer (HBaR) as a means to enhance adversarial robustness. We show that the Hilbert-Schmidt Information bottleneck enhances robustness to adversarial attacks both theoretically and experimentally. In particular, we prove that the HSIC bottleneck regularizer reduces the sensitivity of the classifier to adversarial examples. Our experiments on multiple benchmark datasets and architectures demonstrate that incorporating HBaR attains competitive natural accuracy and improves adversarial robustness, both with and without adversarial examples during training. Finally, we propose a novel framework Pruning-without-Adversarial-training (PwoA) for the purpose of achieving adversarial robustness on resource-constrained systems. PwoA can efficiently prune a previously trained robust neural network while maintaining adversarial robustness, without further generating adversarial examples. We leverage concurrent self-distillation and pruning to preserve knowledge in the original model as well as regularizing the pruned model via the Hilbert-Schmidt Information Bottleneck. We comprehensively evaluate our proposed framework and show its superior performance in terms of both adversarial robustness and efficiency when pruning architectures trained on the benchmark datasets against five state-of-the-art attacks.--Author's abstract

Adversarial robustness is critical for deep learning models to defend against adversarial attacks. Although adversarial training is considered to be one of the most effective ways to improve the model’s adversarial robustness, it usually yields models with lower natural accuracy. In this paper, we argue that, for the attackable examples, traditional adversarial training which utilizes a fixed size perturbation ball can create adversarial examples that deviate far away from the original class towards the target class. Thus, the model’s performance on the natural target class will drop drastically, which leads to the decline of natural accuracy. To this end, we propose the Data-Adaptive Adversarial Training (DAAT) which adaptively adjusts the perturbation ball to a proper size for each of the natural examples with the help of a natural trained calibration network. Besides, a dynamic training strategy empowers the DAAT models with impressive robustness while retaining remarkable natural accuracy. Based on a toy example, we theoretically prove the recession of the natural accuracy caused by adversarial training and show how the data-adaptive perturbation size helps the model resist it. Finally, empirical experiments on benchmark datasets demonstrate the significant improvement of DAAT models on natural accuracy compared with strong baselines.KeywordsAdversarial trainingAdversarial attackAdversarial robustness