Qi Li

Hai Huang, Zhikun Zhang, Yun Shen et al.

Existing studies on neural architecture search (NAS) mainly focus on efficiently and effectively searching for network architectures with better performance. Little progress has been made to systematically understand if the NAS-searched architectures are robust to privacy attacks while abundant work has already shown that human-designed architectures are prone to privacy attacks. In this paper, we fill this gap and systematically measure the privacy risks of NAS architectures. Leveraging the insights from our measurement study, we further explore the cell patterns of cell-based NAS architectures and evaluate how the cell patterns affect the privacy risks of NAS-searched architectures. Through extensive experiments, we shed light on how to design robust NAS architectures against privacy attacks, and also offer a general methodology to understand the hidden correlation between the NAS-searched architectures and other privacy risks.

Bo Yan, Cheng Yang, Chuan Shi et al.

The explosive growth of cyber attacks nowadays, such as malware, spam, and intrusions, caused severe consequences on society. Securing cyberspace has become an utmost concern for organizations and governments. Traditional Machine Learning (ML) based methods are extensively used in detecting cyber threats, but they hardly model the correlations between real-world cyber entities. In recent years, with the proliferation of graph mining techniques, many researchers investigated these techniques for capturing correlations between cyber entities and achieving high performance. It is imperative to summarize existing graph-based cybersecurity solutions to provide a guide for future studies. Therefore, as a key contribution of this paper, we provide a comprehensive review of graph mining for cybersecurity, including an overview of cybersecurity tasks, the typical graph mining techniques, and the general process of applying them to cybersecurity, as well as various solutions for different cybersecurity tasks. For each task, we probe into relevant methods and highlight the graph types, graph approaches, and task levels in their modeling. Furthermore, we collect open datasets and toolkits for graph-based cybersecurity. Finally, we outlook the potential directions of this field for future research.

Sibo Yi, Yule Liu, Zhen Sun et al.

Large Language Models (LLMs) have performed exceptionally in various text-generative tasks, including question answering, translation, code completion, etc. However, the over-assistance of LLMs has raised the challenge of "jailbreaking", which induces the model to generate malicious responses against the usage policy and society by designing adversarial prompts. With the emergence of jailbreak attack methods exploiting different vulnerabilities in LLMs, the corresponding safety alignment measures are also evolving. In this paper, we propose a comprehensive and detailed taxonomy of jailbreak attack and defense methods. For instance, the attack methods are divided into black-box and white-box attacks based on the transparency of the target model. Meanwhile, we classify defense methods into prompt-level and model-level defenses. Additionally, we further subdivide these attack and defense methods into distinct sub-classes and present a coherent diagram illustrating their relationships. We also conduct an investigation into the current evaluation methods and compare them from different perspectives. Our findings aim to inspire future research and practical implementations in safeguarding LLMs against adversarial attacks. Above all, although jailbreak remains a significant concern within the community, we believe that our work enhances the understanding of this domain and provides a foundation for developing more secure LLMs.

Zhengyu Zhao, Hanwei Zhang, Renjue Li et al.

Transferable adversarial images raise critical security concerns for computer vision systems in real-world, black-box attack scenarios. Although many transfer attacks have been proposed, existing research lacks a systematic and comprehensive evaluation. In this paper, we systemize transfer attacks into five categories around the general machine learning pipeline and provide the first comprehensive evaluation, with 23 representative attacks against 11 representative defenses, including the recent, transfer-oriented defense and the real-world Google Cloud Vision. In particular, we identify two main problems of existing evaluations: (1) for attack transferability, lack of intra-category analyses with fair hyperparameter settings, and (2) for attack stealthiness, lack of diverse measures. Our evaluation results validate that these problems have indeed caused misleading conclusions and missing points, and addressing them leads to new, \textit{consensus-challenging} insights, such as (1) an early attack, DI, even outperforms all similar follow-up ones, (2) the state-of-the-art (white-box) defense, DiffPure, is even vulnerable to (black-box) transfer attacks, and (3) even under the same $L_p$ constraint, different attacks yield dramatically different stealthiness results regarding diverse imperceptibility metrics, finer-grained measures, and a user study. We hope that our analyses will serve as guidance on properly evaluating transferable adversarial images and advance the design of attacks and defenses. Code is available at https://github.com/ZhengyuZhao/TransferAttackEval.

Jiating Ma, Yipeng Zhou, Qi Li et al.

To preserve the data privacy, the federated learning (FL) paradigm emerges in which clients only expose model gradients rather than original data for conducting model training. To enhance the protection of model gradients in FL, differentially private federated learning (DPFL) is proposed which incorporates differentially private (DP) noises to obfuscate gradients before they are exposed. Yet, an essential but largely overlooked problem in DPFL is the heterogeneity of clients' privacy requirement, which can vary significantly between clients and extremely complicates the client selection problem in DPFL. In other words, both the data quality and the influence of DP noises should be taken into account when selecting clients. To address this problem, we conduct convergence analysis of DPFL under heterogeneous privacy, a generic client selection strategy, popular DP mechanisms and convex loss. Based on convergence analysis, we formulate the client selection problem to minimize the value of loss function in DPFL with heterogeneous privacy, which is a convex optimization problem and can be solved efficiently. Accordingly, we propose the DPFL-BCS (biased client selection) algorithm. The extensive experiment results with real datasets under both convex and non-convex loss functions indicate that DPFL-BCS can remarkably improve model utility compared with the SOTA baselines.

Zesen Liu, Tianshuo Cong, Xinlei He et al.

Large Language Models (LLMs) excel in various applications, including text generation and complex tasks. However, the misuse of LLMs raises concerns about the authenticity and ethical implications of the content they produce, such as deepfake news, academic fraud, and copyright infringement. Watermarking techniques, which embed identifiable markers in machine-generated text, offer a promising solution to these issues by allowing for content verification and origin tracing. Unfortunately, the robustness of current LLM watermarking schemes under potential watermark removal attacks has not been comprehensively explored. In this paper, to fill this gap, we first systematically comb the mainstream watermarking schemes and removal attacks on machine-generated texts, and then we categorize them into pre-text (before text generation) and post-text (after text generation) classes so that we can conduct diversified analyses. In our experiments, we evaluate eight watermarks (five pre-text, three post-text) and twelve attacks (two pre-text, ten post-text) across 87 scenarios. Evaluation results indicate that (1) KGW and Exponential watermarks offer high text quality and watermark retention but remain vulnerable to most attacks; (2) Post-text attacks are found to be more efficient and practical than pre-text attacks; (3) Pre-text watermarks are generally more imperceptible, as they do not alter text fluency, unlike post-text watermarks; (4) Additionally, combined attack methods can significantly increase effectiveness, highlighting the need for more robust watermarking solutions. Our study underscores the vulnerabilities of current techniques and the necessity for developing more resilient schemes.

Qi Li, Liangzhi Li, Zhouqiang Jiang et al.

Visual prompting, an efficient method for transfer learning, has shown its potential in vision tasks. However, previous works focus exclusively on VP from standard source models, it is still unknown how it performs under the scenario of a robust source model: Whether a visual prompt derived from a robust model can inherit the robustness while suffering from the generalization performance decline, albeit for a downstream dataset that is different from the source dataset? In this work, we get an affirmative answer of the above question and give an explanation on the visual representation level. Moreover, we introduce a novel technique named Prompt Boundary Loose (PBL) to effectively mitigates the suboptimal results of visual prompt on standard accuracy without losing (or even significantly improving) its adversarial robustness when using a robust model as source model. Extensive experiments across various datasets show that our findings are universal and demonstrate the significant benefits of our proposed method.

Minfang Lu, Yuchen Jiang, Huihui Dong et al.

In large-scale industrial e-commerce, the efficiency of an online recommendation system is crucial in delivering highly relevant item/content advertising that caters to diverse business scenarios. However, most existing studies focus solely on item advertising, neglecting the significance of content advertising. This oversight results in inconsistencies within the multi-entity structure and unfair retrieval. Furthermore, the challenge of retrieving top-k advertisements from multi-entity advertisements across different domains adds to the complexity. Recent research proves that user-entity behaviors within different domains exhibit characteristics of differentiation and homogeneity. Therefore, the multi-domain matching models typically rely on the hybrid-experts framework with domain-invariant and domain-specific representations. Unfortunately, most approaches primarily focus on optimizing the combination mode of different experts, failing to address the inherent difficulty in optimizing the expert modules themselves. The existence of redundant information across different domains introduces interference and competition among experts, while the distinct learning objectives of each domain lead to varying optimization challenges among experts. To tackle these issues, we propose robust representation learning for the unified online top-k recommendation. Our approach constructs unified modeling in entity space to ensure data fairness. The robust representation learning employs domain adversarial learning and multi-view wasserstein distribution learning to learn robust representations. Moreover, the proposed method balances conflicting objectives through the homoscedastic uncertainty weights and orthogonality constraints. Various experiments validate the effectiveness and rationality of our proposed method, which has been successfully deployed online to serve real business scenarios.

Rihan Chen, Bin Liu, Han Zhu et al.

Model-based methods for recommender systems have been studied extensively for years. Modern recommender systems usually resort to 1) representation learning models which define user-item preference as the distance between their embedding representations, and 2) embedding-based Approximate Nearest Neighbor (ANN) search to tackle the efficiency problem introduced by large-scale corpus. While providing efficient retrieval, the embedding-based retrieval pattern also limits the model capacity since the form of user-item preference measure is restricted to the distance between their embedding representations. However, for other more precise user-item preference measures, e.g., preference scores directly derived from a deep neural network, they are computationally intractable because of the lack of an efficient retrieval method, and an exhaustive search for all user-item pairs is impractical. In this paper, we propose a novel method to extend ANN search to arbitrary matching functions, e.g., a deep neural network. Our main idea is to perform a greedy walk with a matching function in a similarity graph constructed from all items. To solve the problem that the similarity measures of graph construction and user-item matching function are heterogeneous, we propose a pluggable adversarial training task to ensure the graph search with arbitrary matching function can achieve fairly high precision. Experimental results in both open source and industry datasets demonstrate the effectiveness of our method. The proposed method has been fully deployed in the Taobao display advertising platform and brings a considerable advertising revenue increase. We also summarize our detailed experiences in deployment in this paper.

Bin Liu, Yue Cao, Yutong Lin et al.

This paper introduces a negative margin loss to metric learning based few-shot learning methods. The negative margin loss significantly outperforms regular softmax loss, and achieves state-of-the-art accuracy on three standard few-shot classification benchmarks with few bells and whistles. These results are contrary to the common practice in the metric learning field, that the margin is zero or positive. To understand why the negative margin loss performs well for the few-shot classification, we analyze the discriminability of learned features w.r.t different margins for training and novel classes, both empirically and theoretically. We find that although negative margin reduces the feature discriminability for training classes, it may also avoid falsely mapping samples of the same novel class to multiple peaks or clusters, and thus benefit the discrimination of novel classes. Code is available at https://github.com/bl0/negative-margin.few-shot.

Jinzhu Yan, Haotian Xu, Zhuotao Liu et al.

The emerging programmable networks sparked significant research on Intelligent Network Data Plane (INDP), which achieves learning-based traffic analysis at line-speed. Prior art in INDP focus on deploying tree/forest models on the data plane. We observe a fundamental limitation in tree-based INDP approaches: although it is possible to represent even larger tree/forest tables on the data plane, the flow features that are computable on the data plane are fundamentally limited by hardware constraints. In this paper, we present BoS to push the boundaries of INDP by enabling Neural Network (NN) driven traffic analysis at line-speed. Many types of NNs (such as Recurrent Neural Network (RNN), and transformers) that are designed to work with sequential data have advantages over tree-based models, because they can take raw network data as input without complex feature computations on the fly. However, the challenge is significant: the recurrent computation scheme used in RNN inference is fundamentally different from the match-action paradigm used on the network data plane. BoS addresses this challenge by (i) designing a novel data plane friendly RNN architecture that can execute unlimited RNN time steps with limited data plane stages, effectively achieving line-speed RNN inference; and (ii) complementing the on-switch RNN model with an off-switch transformer-based traffic analysis module to further boost the overall performance. We implement a prototype of BoS using a P4 programmable switch as our data plane, and extensively evaluate it over multiple traffic analysis tasks. The results show that BoS outperforms state-of-the-art in both analysis accuracy and scalability.

Qi Tan, Qi Li, Yi Zhao et al.

Federated Learning (FL) trains a black-box and high-dimensional model among different clients by exchanging parameters instead of direct data sharing, which mitigates the privacy leak incurred by machine learning. However, FL still suffers from membership inference attacks (MIA) or data reconstruction attacks (DRA). In particular, an attacker can extract the information from local datasets by constructing DRA, which cannot be effectively throttled by existing techniques, e.g., Differential Privacy (DP). In this paper, we aim to ensure a strong privacy guarantee for FL under DRA. We prove that reconstruction errors under DRA are constrained by the information acquired by an attacker, which means that constraining the transmitted information can effectively throttle DRA. To quantify the information leakage incurred by FL, we establish a channel model, which depends on the upper bound of joint mutual information between the local dataset and multiple transmitted parameters. Moreover, the channel model indicates that the transmitted information can be constrained through data space operation, which can improve training efficiency and the model accuracy under constrained information. According to the channel model, we propose algorithms to constrain the information transmitted in a single round of local training. With a limited number of training rounds, the algorithms ensure that the total amount of transmitted information is limited. Furthermore, our channel model can be applied to various privacy-enhancing techniques (such as DP) to enhance privacy guarantees against DRA. Extensive experiments with real-world datasets validate the effectiveness of our methods.

Yu Yan, Sheng Sun, Zenghao Duan et al.

Current studies have exposed the risk of Large Language Models (LLMs) generating harmful content by jailbreak attacks. However, they overlook that the direct generation of harmful content from scratch is more difficult than inducing LLM to calibrate benign content into harmful forms. In our study, we introduce a novel attack framework that exploits AdVersArial meTAphoR (AVATAR) to induce the LLM to calibrate malicious metaphors for jailbreaking. Specifically, to answer harmful queries, AVATAR adaptively identifies a set of benign but logically related metaphors as the initial seed. Then, driven by these metaphors, the target LLM is induced to reason and calibrate about the metaphorical content, thus jailbroken by either directly outputting harmful responses or calibrating residuals between metaphorical and professional harmful content. Experimental results demonstrate that AVATAR can effectively and transferable jailbreak LLMs and achieve a state-of-the-art attack success rate across multiple advanced LLMs.

Qing Wang, Yuepei Li, Qiao Qiao et al.

Open Relation Extraction (OpenRE) seeks to identify and extract novel relational facts between named entities from unlabeled data without pre-defined relation schemas. Traditional OpenRE methods typically assume that the unlabeled data consists solely of novel relations or is pre-divided into known and novel instances. However, in real-world scenarios, novel relations are arbitrarily distributed. In this paper, we propose a generalized OpenRE setting that considers unlabeled data as a mixture of both known and novel instances. To address this, we propose MixORE, a two-phase framework that integrates relation classification and clustering to jointly learn known and novel relations. Experiments on three benchmark datasets demonstrate that MixORE consistently outperforms competitive baselines in known relation classification and novel relation clustering. Our findings contribute to the advancement of generalized OpenRE research and real-world applications.

Sibo Yi, Tianshuo Cong, Xinlei He et al.

Small language models (SLMs) have become increasingly prominent in the deployment on edge devices due to their high efficiency and low computational cost. While researchers continue to advance the capabilities of SLMs through innovative training strategies and model compression techniques, the security risks of SLMs have received considerably less attention compared to large language models (LLMs).To fill this gap, we provide a comprehensive empirical study to evaluate the security performance of 13 state-of-the-art SLMs under various jailbreak attacks. Our experiments demonstrate that most SLMs are quite susceptible to existing jailbreak attacks, while some of them are even vulnerable to direct harmful prompts.To address the safety concerns, we evaluate several representative defense methods and demonstrate their effectiveness in enhancing the security of SLMs. We further analyze the potential security degradation caused by different SLM techniques including architecture compression, quantization, knowledge distillation, and so on. We expect that our research can highlight the security challenges of SLMs and provide valuable insights to future work in developing more robust and secure SLMs.

Yonas Sium, Qi Li

Graph Neural Networks (GNNs) have become the leading approach for addressing graph analytical problems in various real-world scenarios. However, GNNs may produce biased predictions against certain demographic subgroups due to node attributes and neighbors surrounding a node. Most current research on GNN fairness focuses predominantly on debiasing GNNs using oversimplified fairness evaluation metrics, which can give a misleading impression of fairness. Understanding the potential evaluation paradoxes due to the complicated nature of the graph structure is crucial for developing effective GNN debiasing mechanisms. In this paper, we examine the effectiveness of current GNN debiasing methods in terms of unfairness evaluation. Specifically, we introduce a community-level strategy to measure bias in GNNs and evaluate debiasing methods at this level. Further, We introduce ComFairGNN, a novel framework designed to mitigate community-level bias in GNNs. Our approach employs a learnable coreset-based debiasing function that addresses bias arising from diverse local neighborhood distributions during GNNs neighborhood aggregation. Comprehensive evaluations on three benchmark datasets demonstrate our model's effectiveness in both accuracy and fairness metrics.

Min Ren, Yunlong Wang, Yuhao Zhu et al.

Insect production for food and feed presents a promising supplement to ensure food safety and address the adverse impacts of agriculture on climate and environment in the future. However, optimisation is required for insect production to realise its full potential. This can be by targeted improvement of traits of interest through selective breeding, an approach which has so far been underexplored and underutilised in insect farming. Here we present a comprehensive review of the selective breeding framework in the context of insect production. We systematically evaluate adjustments of selective breeding techniques to the realm of insects and highlight the essential components integral to the breeding process. The discussion covers every step of a conventional breeding scheme, such as formulation of breeding objectives, phenotyping, estimation of genetic parameters and breeding values, selection of appropriate breeding strategies, and mitigation of issues associated with genetic diversity depletion and inbreeding. This review combines knowledge from diverse disciplines, bridging the gap between animal breeding, quantitative genetics, evolutionary biology, and entomology, offering an integrated view of the insect breeding research area and uniting knowledge which has previously remained scattered across diverse fields of expertise.

Baolin Zheng, Peipei Jiang, Qian Wang et al.

Adversarial attacks against commercial black-box speech platforms, including cloud speech APIs and voice control devices, have received little attention until recent years. The current "black-box" attacks all heavily rely on the knowledge of prediction/confidence scores to craft effective adversarial examples, which can be intuitively defended by service providers without returning these messages. In this paper, we propose two novel adversarial attacks in more practical and rigorous scenarios. For commercial cloud speech APIs, we propose Occam, a decision-only black-box adversarial attack, where only final decisions are available to the adversary. In Occam, we formulate the decision-only AE generation as a discontinuous large-scale global optimization problem, and solve it by adaptively decomposing this complicated problem into a set of sub-problems and cooperatively optimizing each one. Our Occam is a one-size-fits-all approach, which achieves 100% success rates of attacks with an average SNR of 14.23dB, on a wide range of popular speech and speaker recognition APIs, including Google, Alibaba, Microsoft, Tencent, iFlytek, and Jingdong, outperforming the state-of-the-art black-box attacks. For commercial voice control devices, we propose NI-Occam, the first non-interactive physical adversarial attack, where the adversary does not need to query the oracle and has no access to its internal information and training data. We combine adversarial attacks with model inversion attacks, and thus generate the physically-effective audio AEs with high transferability without any interaction with target devices. Our experimental results show that NI-Occam can successfully fool Apple Siri, Microsoft Cortana, Google Assistant, iFlytek and Amazon Echo with an average SRoA of 52% and SNR of 9.65dB, shedding light on non-interactive physical attacks against voice control devices.

Tao Liu, Bowen Yang, Qi Li et al.

Information flows are intrinsic properties of an multi-stage manufacturing systems (MMS). In computer security, a basic information flow tracking technique is dynamic taint analysis (DTA). DTA tracks taint propagation from one data variable (e.g., a buffer holding a HTTP request) to another. Taint propagation paths are typically determined by data flows and implicit flows in a computer program. And the union of all the taint propagation paths forms a taint graph. It is clear that taints graphs could significantly enhance intrusion diagnosis. However, the existing DTA techniques cannot be directly used in an MMS, and a main reason is as follows: Without manufacturing-specific taint propagation rules, DTA cannot be implemented. In this work, we conduct a case study which (a) extends the existing DTA method with manufacturing-specific taint propagation rules, and (b) applies the extended method to perform preliminary intrusion diagnosis with a small-scale test-bed.

Yinwei Wei, Xiang Wang, Qi Li et al.

Recommending cold-start items is a long-standing and fundamental challenge in recommender systems. Without any historical interaction on cold-start items, CF scheme fails to use collaborative signals to infer user preference on these items. To solve this problem, extensive studies have been conducted to incorporate side information into the CF scheme. Specifically, they employ modern neural network techniques (e.g., dropout, consistency constraint) to discover and exploit the coalition effect of content features and collaborative representations. However, we argue that these works less explore the mutual dependencies between content features and collaborative representations and lack sufficient theoretical supports, thus resulting in unsatisfactory performance. In this work, we reformulate the cold-start item representation learning from an information-theoretic standpoint. It aims to maximize the mutual dependencies between item content and collaborative signals. Specifically, the representation learning is theoretically lower-bounded by the integration of two terms: mutual information between collaborative embeddings of users and items, and mutual information between collaborative embeddings and feature representations of items. To model such a learning process, we devise a new objective function founded upon contrastive learning and develop a simple yet effective Contrastive Learning-based Cold-start Recommendation framework(CLCRec). In particular, CLCRec consists of three components: contrastive pair organization, contrastive embedding, and contrastive optimization modules. It allows us to preserve collaborative signals in the content representations for both warm and cold-start items. Through extensive experiments on four publicly accessible datasets, we observe that CLCRec achieves significant improvements over state-of-the-art approaches in both warm- and cold-start scenarios.

Chuanpu Fu, Qi Li, Meng Shen et al.

Machine learning (ML) based malicious traffic detection is an emerging security paradigm, particularly for zero-day attack detection, which is complementary to existing rule based detection. However, the existing ML based detection has low detection accuracy and low throughput incurred by inefficient traffic features extraction. Thus, they cannot detect attacks in realtime especially in high throughput networks. Particularly, these detection systems similar to the existing rule based detection can be easily evaded by sophisticated attacks. To this end, we propose Whisper, a realtime ML based malicious traffic detection system that achieves both high accuracy and high throughput by utilizing frequency domain features. It utilizes sequential features represented by the frequency domain features to achieve bounded information loss, which ensures high detection accuracy, and meanwhile constrains the scale of features to achieve high detection throughput. Particularly, attackers cannot easily interfere with the frequency domain features and thus Whisper is robust against various evasion attacks. Our experiments with 42 types of attacks demonstrate that, compared with the state-of-theart systems, Whisper can accurately detect various sophisticated and stealthy attacks, achieving at most 18.36% improvement, while achieving two orders of magnitude throughput. Even under various evasion attacks, Whisper is still able to maintain around 90% detection accuracy.

Guangmeng Zhou, Ke Xu, Qi Li et al.

In multi-party collaborative learning, the parameter server sends a global model to each data holder for local training and then aggregates committed models globally to achieve privacy protection. However, both the dragger issue of synchronous collaborative learning and the staleness issue of asynchronous collaborative learning make collaborative learning inefficient in real-world heterogeneous environments. We propose a novel and efficient collaborative learning framework named AdaptCL, which generates an adaptive sub-model dynamically from the global base model for each data holder, without any prior information about worker capability. All workers (data holders) achieve approximately identical update time as the fastest worker by equipping them with capability-adapted pruned models. Thus the training process can be dramatically accelerated. Besides, we tailor the efficient pruned rate learning algorithm and pruning approach for AdaptCL. Meanwhile, AdaptCL provides a mechanism for handling the trade-off between accuracy and time overhead and can be combined with other techniques to accelerate training further. Empirical results show that AdaptCL introduces little computing and communication overhead. AdaptCL achieves time savings of more than 41\% on average and improves accuracy in a low heterogeneous environment. In a highly heterogeneous environment, AdaptCL achieves a training speedup of 6.2x with a slight loss of accuracy.

Dian Chen, Hongxin Hu, Qian Wang et al.

Transfer learning eases the burden of training a well-performed model from scratch, especially when training data is scarce and computation power is limited. In deep learning, a typical strategy for transfer learning is to freeze the early layers of a pre-trained model and fine-tune the rest of its layers on the target domain. Previous work focuses on the accuracy of the transferred model but neglects the transfer of adversarial robustness. In this work, we first show that transfer learning improves the accuracy on the target domain but degrades the inherited robustness of the target model. To address such a problem, we propose a novel cooperative adversarially-robust transfer learning (CARTL) by pre-training the model via feature distance minimization and fine-tuning the pre-trained model with non-expansive fine-tuning for target domain tasks. Empirical results show that CARTL improves the inherited robustness by about 28% at most compared with the baseline with the same degree of accuracy. Furthermore, we study the relationship between the batch normalization (BN) layers and the robustness in the context of transfer learning, and we reveal that freezing BN layers can further boost the robustness transfer.

Fengting Li, Xuankai Liu, Xiaoli Zhang et al.

Deep neural networks (DNNs) have been applied in a wide range of applications,e.g.,face recognition and image classification; however,they are vulnerable to adversarial examples. By adding a small amount of imperceptible perturbations,an attacker can easily manipulate the outputs of a DNN. Particularly,the localized adversarial examples only perturb a small and contiguous region of the target object,so that they are robust and effective in both digital and physical worlds. Although the localized adversarial examples have more severe real-world impacts than traditional pixel attacks,they have not been well addressed in the literature. In this paper,we propose a generic defense system called TaintRadar to accurately detect localized adversarial examples via analyzing critical regions that have been manipulated by attackers. The main idea is that when removing critical regions from input images,the ranking changes of adversarial labels will be larger than those of benign labels. Compared with existing defense solutions,TaintRadar can effectively capture sophisticated localized partial attacks, e.g.,the eye-glasses attack,while not requiring additional training or fine-tuning of the original model's structure. Comprehensive experiments have been conducted in both digital and physical worlds to verify the effectiveness and robustness of our defense.

Meng Shen, Hao Yu, Liehuang Zhu et al.

Deep neural networks (DNNs) have been increasingly used in face recognition (FR) systems. Recent studies, however, show that DNNs are vulnerable to adversarial examples, which can potentially mislead the FR systems using DNNs in the physical world. Existing attacks on these systems either generate perturbations working merely in the digital world, or rely on customized equipments to generate perturbations and are not robust in varying physical environments. In this paper, we propose FaceAdv, a physical-world attack that crafts adversarial stickers to deceive FR systems. It mainly consists of a sticker generator and a transformer, where the former can craft several stickers with different shapes and the latter transformer aims to digitally attach stickers to human faces and provide feedbacks to the generator to improve the effectiveness of stickers. We conduct extensive experiments to evaluate the effectiveness of FaceAdv on attacking 3 typical FR systems (i.e., ArcFace, CosFace and FaceNet). The results show that compared with a state-of-the-art attack, FaceAdv can significantly improve success rate of both dodging and impersonating attacks. We also conduct comprehensive evaluations to demonstrate the robustness of FaceAdv.

Lingchen Zhao, Qian Wang, Cong Wang et al.

Machine Learning as a Service (MLaaS) allows clients with limited resources to outsource their expensive ML tasks to powerful servers. Despite the huge benefits, current MLaaS solutions still lack strong assurances on: 1) service correctness (i.e., whether the MLaaS works as expected); 2) trustworthy accounting (i.e., whether the bill for the MLaaS resource consumption is correctly accounted); 3) fair payment (i.e., whether a client gets the entire MLaaS result before making the payment). Without these assurances, unfaithful service providers can return improperly-executed ML task results or partially trained ML models while asking for over-claimed rewards. Moreover, it is hard to argue for wide adoption of MLaaS to both the client and the service provider, especially in the open market without a trusted third party. In this paper, we present VeriML, a novel and efficient framework to bring integrity assurances and fair payments to MLaaS. With VeriML, clients can be assured that ML tasks are correctly executed on an untrusted server and the resource consumption claimed by the service provider equals to the actual workload. We strategically use succinct non-interactive arguments of knowledge (SNARK) on randomly-selected iterations during the ML training phase for efficiency with tunable probabilistic assurance. We also develop multiple ML-specific optimizations to the arithmetic circuit required by SNARK. Our system implements six common algorithms: linear regression, logistic regression, neural network, support vector machine, Kmeans and decision tree. The experimental results have validated the practical performance of VeriML.

Michael A. Alcorn, Qi Li, Zhitao Gong et al.

Despite excellent performance on stationary test sets, deep neural networks (DNNs) can fail to generalize to out-of-distribution (OoD) inputs, including natural, non-adversarial ones, which are common in real-world settings. In this paper, we present a framework for discovering DNN failures that harnesses 3D renderers and 3D models. That is, we estimate the parameters of a 3D renderer that cause a target DNN to misbehave in response to the rendered image. Using our framework and a self-assembled dataset of 3D objects, we investigate the vulnerability of DNNs to OoD poses of well-known objects in ImageNet. For objects that are readily recognized by DNNs in their canonical poses, DNNs incorrectly classify 97% of their pose space. In addition, DNNs are highly sensitive to slight pose perturbations. Importantly, adversarial poses transfer across models and datasets. We find that 99.9% and 99.4% of the poses misclassified by Inception-v3 also transfer to the AlexNet and ResNet-50 image classifiers trained on the same ImageNet dataset, respectively, and 75.5% transfer to the YOLOv3 object detector trained on MS COCO.

Jinwen Wang, Yueqiang Cheng, Qi Li et al.

Intel has introduced a trusted computing technology, Intel Software Guard Extension (SGX), which provides an isolated and secure execution environment called enclave for a user program without trusting any privilege software (e.g., an operating system or a hypervisor) or firmware. Nevertheless, SGX is vulnerable to several side channel attacks (e.g. page-fault-based attack and cache-based attack). In this paper, we explore a new, yet critical side channel attack in SGX, interface-based side channel attack, which can infer the information of the enclave input data. The root cause of the interface-based side channel attack is the input dependent interface invocation information (e.g., interface information and invocation patterns) which can be observed by the untrusted privilege software can reveal the control flow in the enclave. We study the methodology which can be used to conduct the interface-based side channel attack. To illustrate the effectiveness of the interface-based side-channel attacks, we use our methodology to infer whether tracked web pages have been processed by the SGX-assisted NFV platforms and achieve the accuracy of 87.6% and recall of 76.6%. We also identify the packets which belong to the tracked web pages, with the accuracy of 67.9%and recall of 71.1%. We finally propose some countermeasures to defense the interface-based side channel attack in SGX-assisted applications.

Man Zhou, Qian Wang, Jingxiao Yang et al.

Pattern lock has been widely used for authentication to protect user privacy on mobile devices (e.g., smartphones and tablets). Given its pervasive usage, the compromise of pattern lock could lead to serious consequences. Several attacks have been constructed to crack the lock. However, these approaches require the attackers to either be physically close to the target device or be able to manipulate the network facilities (e.g., WiFi hotspots) used by the victims. Therefore, the effectiveness of the attacks is significantly impacted by the environment of mobile devices. Also, these attacks are not scalable since they cannot easily infer unlock patterns of a large number of devices. Motivated by an observation that fingertip motions on the screen of a mobile device can be captured by analyzing surrounding acoustic signals on it, we propose PatternListener, a novel acoustic attack that cracks pattern lock by analyzing imperceptible acoustic signals reflected by the fingertip. It leverages speakers and microphones of the victim's device to play imperceptible audio and record the acoustic signals reflected by the fingertip. In particular, it infers each unlock pattern by analyzing individual lines that compose the pattern and are the trajectories of the fingertip. We propose several algorithms to construct signal segments according to the captured signals for each line and infer possible candidates of each individual line according to the signal segments. Finally, we map all line candidates into grid patterns and thereby obtain the candidates of the entire unlock pattern. We implement a PatternListener prototype by using off-the-shelf smartphones and thoroughly evaluate it using 130 unique patterns. The real experimental results demonstrate that PatternListener can successfully exploit over 90% patterns within five attempts.