An Adaptive Model Ensemble Adversarial Attack for Boosting Adversarial Transferability

While the transferability property of adversarial examples allows the adversary to perform black-box attacks (i.e., the attacker has no knowledge about the target model), the transfer-based adversarial attacks have gained great attention. Previous works mostly study gradient variation or image transformations to amplify the distortion on critical parts of inputs. These methods can work on transferring across models with limited differences, i.e., from CNNs to CNNs, but always fail in transferring across models with wide differences, such as from CNNs to ViTs. Alternatively, model ensemble adversarial attacks are proposed to fuse outputs from surrogate models with diverse architectures to get an ensemble loss, making the generated adversarial example more likely to transfer to other models as it can fool multiple models concurrently. However, existing ensemble attacks simply fuse the outputs of the surrogate models evenly, thus are not efficacious to capture and amplify the intrinsic transfer information of adversarial examples. In this paper, we propose an adaptive ensemble attack, dubbed AdaEA, to adaptively control the fusion of the outputs from each model, via monitoring the discrepancy ratio of their contributions towards the adversarial objective. Furthermore, an extra disparity-reduced filter is introduced to further synchronize the update direction. As a result, we achieve considerable improvement over the existing ensemble attacks on various datasets, and the proposed AdaEA can also boost existing transfer-based attacks, which further demonstrates its efficacy and versatility.

Fedward: Flexible Federated Backdoor Defense Framework with Non-IID Data

Federated learning (FL) enables multiple clients to collaboratively train deep learning models while considering sensitive local datasets' privacy. However, adversaries can manipulate datasets and upload models by injecting triggers for federated backdoor attacks (FBA). Existing defense strategies against FBA consider specific and limited attacker models, and a sufficient amount of noise to be injected only mitigates rather than eliminates FBA. To address these deficiencies, we introduce a Flexible Federated Backdoor Defense Framework (Fedward) to ensure the elimination of adversarial backdoors. We decompose FBA into various attacks, and design amplified magnitude sparsification (AmGrad) and adaptive OPTICS clustering (AutoOPTICS) to address each attack. Meanwhile, Fedward uses the adaptive clipping method by regarding the number of samples in the benign group as constraints on the boundary. This ensures that Fedward can maintain the performance for the Non-IID scenario. We conduct experimental evaluations over three benchmark datasets and thoroughly compare them to state-of-the-art studies. The results demonstrate the promising defense performance from Fedward, moderately improved by 33% $\sim$ 75 in clustering defense methods, and 96.98%, 90.74%, and 89.8% for Non-IID to the utmost extent for the average FBA success rate over MNIST, FMNIST, and CIFAR10, respectively.

When Evolutionary Computation Meets Privacy

Recently, evolutionary computation (EC) has been promoted by machine learning, distributed computing, and big data technologies, resulting in new research directions of EC like distributed EC and surrogate-assisted EC. These advances have significantly improved the performance and the application scope of EC, but also trigger privacy leakages, such as the leakage of optimal results and surrogate model. Accordingly, evolutionary computation combined with privacy protection is becoming an emerging topic. However, privacy concerns in evolutionary computation lack a systematic exploration, especially for the object, motivation, position, and method of privacy protection. To this end, in this paper, we discuss three typical optimization paradigms (i.e., \textit{centralized optimization, distributed optimization, and data-driven optimization}) to characterize optimization modes of evolutionary computation and propose BOOM to sort out privacy concerns in evolutionary computation. Specifically, the centralized optimization paradigm allows clients to outsource optimization problems to the centralized server and obtain optimization solutions from the server. While the distributed optimization paradigm exploits the storage and computational power of distributed devices to solve optimization problems. Also, the data-driven optimization paradigm utilizes data collected in history to tackle optimization problems lacking explicit objective functions. Particularly, this paper adopts BOOM to characterize the object and motivation of privacy protection in three typical optimization paradigms and discusses potential privacy-preserving technologies balancing optimization performance and privacy guarantees in three typical optimization paradigms. Furthermore, this paper attempts to foresee some new research directions of privacy-preserving evolutionary computation.

SRoUDA: Meta Self-training for Robust Unsupervised Domain Adaptation

As acquiring manual labels on data could be costly, unsupervised domain adaptation (UDA), which transfers knowledge learned from a rich-label dataset to the unlabeled target dataset, is gaining increasing popularity. While extensive studies have been devoted to improving the model accuracy on target domain, an important issue of model robustness is neglected. To make things worse, conventional adversarial training (AT) methods for improving model robustness are inapplicable under UDA scenario since they train models on adversarial examples that are generated by supervised loss function. In this paper, we present a new meta self-training pipeline, named SRoUDA, for improving adversarial robustness of UDA models. Based on self-training paradigm, SRoUDA starts with pre-training a source model by applying UDA baseline on source labeled data and taraget unlabeled data with a developed random masked augmentation (RMA), and then alternates between adversarial target model training on pseudo-labeled target data and finetuning source model by a meta step. While self-training allows the direct incorporation of AT in UDA, the meta step in SRoUDA further helps in mitigating error propagation from noisy pseudo labels. Extensive experiments on various benchmark datasets demonstrate the state-of-the-art performance of SRoUDA where it achieves significant model robustness improvement without harming clean accuracy. Code is available at https://github.com/Vision.

Refiner: Data Refining against Gradient Leakage Attacks in Federated Learning

Federated Learning (FL) is pervasive in privacy-focused IoT environments since it enables avoiding privacy leakage by training models with gradients instead of data. Recent works show the uploaded gradients can be employed to reconstruct data, i.e., gradient leakage attacks, and several defenses are designed to alleviate the risk by tweaking the gradients. However, these defenses exhibit weak resilience against threatening attacks, as the effectiveness builds upon the unrealistic assumptions that deep neural networks are simplified as linear models. In this paper, without such unrealistic assumptions, we present a novel defense, called Refiner, instead of perturbing gradients, which refines ground-truth data to craft robust data that yields sufficient utility but with the least amount of privacy information, and then the gradients of robust data are uploaded. To craft robust data, Refiner promotes the gradients of critical parameters associated with robust data to close ground-truth ones while leaving the gradients of trivial parameters to safeguard privacy. Moreover, to exploit the gradients of trivial parameters, Refiner utilizes a well-designed evaluation network to steer robust data far away from ground-truth data, thereby alleviating privacy leakage risk. Extensive experiments across multiple benchmark datasets demonstrate the superior defense effectiveness of Refiner at defending against state-of-the-art threats.

MaskBlock: Transferable Adversarial Examples with Bayes Approach

The transferability of adversarial examples (AEs) across diverse models is of critical importance for black-box adversarial attacks, where attackers cannot access the information about black-box models. However, crafted AEs always present poor transferability. In this paper, by regarding the transferability of AEs as generalization ability of the model, we reveal that vanilla black-box attacks craft AEs via solving a maximum likelihood estimation (MLE) problem. For MLE, the results probably are model-specific local optimum when available data is small, i.e., limiting the transferability of AEs. By contrast, we re-formulate crafting transferable AEs as the maximizing a posteriori probability estimation problem, which is an effective approach to boost the generalization of results with limited available data. Because Bayes posterior inference is commonly intractable, a simple yet effective method called MaskBlock is developed to approximately estimate. Moreover, we show that the formulated framework is a generalization version for various attack methods. Extensive experiments illustrate MaskBlock can significantly improve the transferability of crafted adversarial examples by up to about 20%.

Defense against Backdoor Attacks via Identifying and Purifying Bad Neurons

The opacity of neural networks leads their vulnerability to backdoor attacks, where hidden attention of infected neurons is triggered to override normal predictions to the attacker-chosen ones. In this paper, we propose a novel backdoor defense method to mark and purify the infected neurons in the backdoored neural networks. Specifically, we first define a new metric, called benign salience. By combining the first-order gradient to retain the connections between neurons, benign salience can identify the infected neurons with higher accuracy than the commonly used metric in backdoor defense. Then, a new Adaptive Regularization (AR) mechanism is proposed to assist in purifying these identified infected neurons via fine-tuning. Due to the ability to adapt to different magnitudes of parameters, AR can provide faster and more stable convergence than the common regularization mechanism in neuron purifying. Extensive experimental results demonstrate that our method can erase the backdoor in neural networks with negligible performance degradation.

Evolution as a Service: A Privacy-Preserving Genetic Algorithm for Combinatorial Optimization

Evolutionary algorithms (EAs), such as the genetic algorithm (GA), offer an elegant way to handle combinatorial optimization problems (COPs). However, limited by expertise and resources, most users do not have enough capability to implement EAs to solve COPs. An intuitive and promising solution is to outsource evolutionary operations to a cloud server, whilst it suffers from privacy concerns. To this end, this paper proposes a novel computing paradigm, evolution as a service (EaaS), where a cloud server renders evolutionary computation services for users without sacrificing users' privacy. Inspired by the idea of EaaS, this paper designs PEGA, a novel privacy-preserving GA for COPs. Specifically, PEGA enables users outsourcing COPs to the cloud server holding a competitive GA and approximating the optimal solution in a privacy-preserving manner. PEGA features the following characteristics. First, any user without expertise and enough resources can solve her COPs. Second, PEGA does not leak contents of optimization problems, i.e., users' privacy. Third, PEGA has the same capability as the conventional GA to approximate the optimal solution. We implements PEGA falling in a twin-server architecture and evaluates it in the traveling salesman problem (TSP, a widely known COP). Particularly, we utilize encryption cryptography to protect users' privacy and carefully design a suit of secure computing protocols to support evolutionary operators of GA on encrypted data. Privacy analysis demonstrates that PEGA does not disclose the contents of the COP to the cloud server. Experimental evaluation results on four TSP datasets show that PEGA is as effective as the conventional GA in approximating the optimal solution.

Case-Aware Adversarial Training

The neural network (NN) becomes one of the most heated type of models in various signal processing applications. However, NNs are extremely vulnerable to adversarial examples (AEs). To defend AEs, adversarial training (AT) is believed to be the most effective method while due to the intensive computation, AT is limited to be applied in most applications. In this paper, to resolve the problem, we design a generic and efficient AT improvement scheme, namely case-aware adversarial training (CAT). Specifically, the intuition stems from the fact that a very limited part of informative samples can contribute to most of model performance. Alternatively, if only the most informative AEs are used in AT, we can lower the computation complexity of AT significantly as maintaining the defense effect. To achieve this, CAT achieves two breakthroughs. First, a method to estimate the information degree of adversarial examples is proposed for AE filtering. Second, to further enrich the information that the NN can obtain from AEs, CAT involves a weight estimation and class-level balancing based sampling strategy to increase the diversity of AT at each iteration. Extensive experiments show that CAT is faster than vanilla AT by up to 3x while achieving competitive defense effect.

Enhance transferability of adversarial examples with model architecture

Transferability of adversarial examples is of critical importance to launch black-box adversarial attacks, where attackers are only allowed to access the output of the target model. However, under such a challenging but practical setting, the crafted adversarial examples are always prone to overfitting to the proxy model employed, presenting poor transferability. In this paper, we suggest alleviating the overfitting issue from a novel perspective, i.e., designing a fitted model architecture. Specifically, delving the bottom of the cause of poor transferability, we arguably decompose and reconstruct the existing model architecture into an effective model architecture, namely multi-track model architecture (MMA). The adversarial examples crafted on the MMA can maximumly relieve the effect of model-specified features to it and toward the vulnerable directions adopted by diverse architectures. Extensive experimental evaluation demonstrates that the transferability of adversarial examples based on the MMA significantly surpass other state-of-the-art model architectures by up to 40% with comparable overhead.