Adversarial training has been demonstrated to be one of the most effective remedies for defending adversarial examples, yet it often suffers from the huge robustness generalization gap on unseen testing adversaries, deemed as the \emph{adversarially robust generalization problem}. Despite the preliminary understandings devoted on adversarially robust generalization, little is known from the architectural perspective. Thus, this paper tries to bridge the gap by systematically examining the most representative architectures (e.g., Vision Transformers and CNNs). In particular, we first comprehensively evaluated \emph{20} adversarially trained architectures on ImageNette and CIFAR-10 datasets towards several adversaries (multiple $\ell_p$-norm adversarial attacks), and found that Vision Transformers (e.g., PVT, CoAtNet) often yield better adversarially robust generalization. To further understand what architectural ingredients favor adversarially robust generalization, we delve into several key building blocks and revealed the fact via the lens of Rademacher complexity that the higher weight sparsity contributes significantly towards the better adversarially robust generalization of Vision Transformers, which can be often achieved by attention layers. Our extensive studies discovered the close relationship between architectural design and adversarially robust generalization, and instantiated several important insights. We hope our findings could help to better understand the mechanism towards designing robust deep learning architectures.
Knowledge distillation (KD) has shown its effectiveness for object detection, where it trains a compact object detector under the supervision of both AI knowledge (teacher detector) and human knowledge (human expert). However, existing studies treat the AI knowledge and human knowledge consistently and adopt a uniform data augmentation strategy during learning, which would lead to the biased learning of multi-scale objects and insufficient learning for the teacher detector causing unsatisfactory distillation performance. To tackle these problems, we propose the sample-specific data augmentation and adversarial feature augmentation. Firstly, to mitigate the impact incurred by multi-scale objects, we propose an adaptive data augmentation based on our observations from the Fourier perspective. Secondly, we propose a feature augmentation method based on adversarial examples for better mimicking AI knowledge to make up for the insufficient information mining of the teacher detector. Furthermore, our proposed method is unified and easily extended to other KD methods. Extensive experiments demonstrate the effectiveness of our framework and improve the performance of state-of-the-art methods in one-stage and two-stage detectors, bringing at most 0.5 mAP gains.
Adversarial training (AT) methods are effective against adversarial attacks, yet they introduce severe disparity of accuracy and robustness between different classes, known as the robust fairness problem. Previously proposed Fair Robust Learning (FRL) adaptively reweights different classes to improve fairness. However, the performance of the better-performed classes decreases, leading to a strong performance drop. In this paper, we observed two unfair phenomena during adversarial training: different difficulties in generating adversarial examples from each class (source-class fairness) and disparate target class tendencies when generating adversarial examples (target-class fairness). From the observations, we propose Balance Adversarial Training (BAT) to address the robust fairness problem. Regarding source-class fairness, we adjust the attack strength and difficulties of each class to generate samples near the decision boundary for easier and fairer model learning; considering target-class fairness, by introducing a uniform distribution constraint, we encourage the adversarial example generation process for each class with a fair tendency. Extensive experiments conducted on multiple datasets (CIFAR-10, CIFAR-100, and ImageNette) demonstrate that our method can significantly outperform other baselines in mitigating the robust fairness problem (+5-10\% on the worst class accuracy)
Extensive evidence has demonstrated that deep neural networks (DNNs) are vulnerable to backdoor attacks, which motivates the development of backdoor detection methods. Existing backdoor detection methods are typically tailored for backdoor attacks with individual specific types (e.g., patch-based or perturbation-based). However, adversaries are likely to generate multiple types of backdoor attacks in practice, which challenges the current detection strategies. Based on the fact that adversarial perturbations are highly correlated with trigger patterns, this paper proposes the Adaptive Perturbation Generation (APG) framework to detect multiple types of backdoor attacks by adaptively injecting adversarial perturbations. Since different trigger patterns turn out to show highly diverse behaviors under the same adversarial perturbations, we first design the global-to-local strategy to fit the multiple types of backdoor triggers via adjusting the region and budget of attacks. To further increase the efficiency of perturbation injection, we introduce a gradient-guided mask generation strategy to search for the optimal regions for adversarial attacks. Extensive experiments conducted on multiple datasets (CIFAR-10, GTSRB, Tiny-ImageNet) demonstrate that our method outperforms state-of-the-art baselines by large margins(+12%).
Billions of people are sharing their daily life images on social media every day. However, their biometric information (e.g., fingerprint) could be easily stolen from these images. The threat of fingerprint leakage from social media raises a strong desire for anonymizing shared images while maintaining image qualities, since fingerprints act as a lifelong individual biometric password. To guard the fingerprint leakage, adversarial attack emerges as a solution by adding imperceptible perturbations on images. However, existing works are either weak in black-box transferability or appear unnatural. Motivated by visual perception hierarchy (i.e., high-level perception exploits model-shared semantics that transfer well across models while low-level perception extracts primitive stimulus and will cause high visual sensitivities given suspicious stimulus), we propose FingerSafe, a hierarchical perceptual protective noise injection framework to address the mentioned problems. For black-box transferability, we inject protective noises on fingerprint orientation field to perturb the model-shared high-level semantics (i.e., fingerprint ridges). Considering visual naturalness, we suppress the low-level local contrast stimulus by regularizing the response of Lateral Geniculate Nucleus. Our FingerSafe is the first to provide feasible fingerprint protection in both digital (up to 94.12%) and realistic scenarios (Twitter and Facebook, up to 68.75%). Our code can be found at https://github.com/nlsde-safety-team/FingerSafe.
To operate in real-world high-stakes environments, deep learning systems have to endure noises that have been continuously thwarting their robustness. Data-end defense, which improves robustness by operations on input data instead of modifying models, has attracted intensive attention due to its feasibility in practice. However, previous data-end defenses show low generalization against diverse noises and weak transferability across multiple models. Motivated by the fact that robust recognition depends on both local and global features, we propose a defensive patch generation framework to address these problems by helping models better exploit these features. For the generalization against diverse noises, we inject class-specific identifiable patterns into a confined local patch prior, so that defensive patches could preserve more recognizable features towards specific classes, leading models for better recognition under noises. For the transferability across multiple models, we guide the defensive patches to capture more global feature correlations within a class, so that they could activate model-shared global perceptions and transfer better among models. Our defensive patches show great potentials to improve application robustness in practice by simply sticking them around target objects. Extensive experiments show that we outperform others by large margins (improve 20+\% accuracy for both adversarial and corruption robustness on average in the digital and physical world). Our codes are available at https://github.com/nlsde-safety-team/DefensivePatch
The large pre-trained BERT has achieved remarkable performance on Natural Language Processing (NLP) tasks but is also computation and memory expensive. As one of the powerful compression approaches, binarization extremely reduces the computation and memory consumption by utilizing 1-bit parameters and bitwise operations. Unfortunately, the full binarization of BERT (i.e., 1-bit weight, embedding, and activation) usually suffer a significant performance drop, and there is rare study addressing this problem. In this paper, with the theoretical justification and empirical analysis, we identify that the severe performance drop can be mainly attributed to the information degradation and optimization direction mismatch respectively in the forward and backward propagation, and propose BiBERT, an accurate fully binarized BERT, to eliminate the performance bottlenecks. Specifically, BiBERT introduces an efficient Bi-Attention structure for maximizing representation information statistically and a Direction-Matching Distillation (DMD) scheme to optimize the full binarized BERT accurately. Extensive experiments show that BiBERT outperforms both the straightforward baseline and existing state-of-the-art quantized BERTs with ultra-low bit activations by convincing margins on the NLP benchmark. As the first fully binarized BERT, our method yields impressive 56.3 times and 31.2 times saving on FLOPs and model size, demonstrating the vast advantages and potential of the fully binarized BERT model in real-world resource-constrained scenarios.
Crowd counting, which is significantly important for estimating the number of people in safety-critical scenes, has been shown to be vulnerable to adversarial examples in the physical world (e.g., adversarial patches). Though harmful, adversarial examples are also valuable for assessing and better understanding model robustness. However, existing adversarial example generation methods in crowd counting scenarios lack strong transferability among different black-box models. Motivated by the fact that transferability is positively correlated to the model-invariant characteristics, this paper proposes the Perceptual Adversarial Patch (PAP) generation framework to learn the shared perceptual features between models by exploiting both the model scale perception and position perception. Specifically, PAP exploits differentiable interpolation and density attention to help learn the invariance between models during training, leading to better transferability. In addition, we surprisingly found that our adversarial patches could also be utilized to benefit the performance of vanilla models for alleviating several challenges including cross datasets and complex backgrounds. Extensive experiments under both digital and physical world scenarios demonstrate the effectiveness of our PAP.
Deep neural networks (DNNs) are vulnerable to adversarial noises, which motivates the benchmark of model robustness. Existing benchmarks mainly focus on evaluating the defenses, but there are no comprehensive studies of how architecture design and general training techniques affect robustness. Comprehensively benchmarking their relationships will be highly beneficial for better understanding and developing robust DNNs. Thus, we propose RobustART, the first comprehensive Robustness investigation benchmark on ImageNet (including open-source toolkit, pre-trained model zoo, datasets, and analyses) regarding ARchitecture design (44 human-designed off-the-shelf architectures and 1200+ networks from neural architecture search) and Training techniques (10+ general techniques, e.g., data augmentation) towards diverse noises (adversarial, natural, and system noises). Extensive experiments revealed and substantiated several insights for the first time, for example: (1) adversarial training largely improves the clean accuracy and all types of robustness for Transformers and MLP-Mixers; (2) with comparable sizes, CNNs > Transformers > MLP-Mixers on robustness against natural and system noises; Transformers > MLP-Mixers > CNNs on adversarial robustness; (3) for some light-weight architectures (e.g., EfficientNet, MobileNetV2, and MobileNetV3), increasing model sizes or using extra training data cannot improve robustness. Our benchmark http://robust.art/ : (1) presents an open-source platform for conducting comprehensive evaluation on diverse robustness types; (2) provides a variety of pre-trained models with different training techniques to facilitate robustness evaluation; (3) proposes a new view to better understand the mechanism towards designing robust DNN architectures, backed up by the analysis. We will continuously contribute to building this ecosystem for the community.