The Deep Neural Network (DNN) based detection model’s dependency on third-party crowdsourced sources poses a new security threat from backdoor attacks against malware detectors. Attackers attempt to inject hidden backdoors into the target model, allowing it to perform well on clean samples. Once the attacker-defined trigger activates the hidden backdoor, the model predictions for poisoned samples are maliciously altered. Different from existing backdoor attacks either utilize fixed triggers or generate sample-specific triggers, we explore a novel backdoor attack paradigm in malware domain and propose MalPDT, in which backdoor triggers achieve dynamic variability in trigger patterns and retain compatibility across malware samples. We train a generator capable of hiding information to produce dynamically variable encoded byte segments, which are then injected as triggers into the unused regions of PE malware in a functionality-preserving manner. In MalPDT, any combination of a malware sample and a trigger can form a poisoned sample capable of activating the backdoor, enabling plug-and-play capability. We conduct extensive experiments to validate the effectiveness of MalPDT in attacking models with or without defenses.

The adversarial example presents new security threats to trustworthy detection systems. In the context of evading dynamic detection based on API call sequences, a practical approach involves inserting perturbing API calls to modify these sequences. The type of inserted API calls and their insertion locations are crucial for generating an effective adversarial API call sequence. Existing methods either optimize the inserted API calls while neglecting the insertion positions or treat these optimizations as separate processes. This can lead to inefficient attacks that insert a large number of unnecessary API calls. To address this issue, we propose a novel reinforcement learning (RL) framework, dubbed GAME-RL, which simultaneously optimizes both the perturbing APIs and their insertion positions. Specifically, we define malware modification through IAT (Import Address Table) hooking as a sequential decision-making process. We introduce an invalid action masking and an auto-regressive policy head within the RL framework, ensuring the feasibility of IAT hooking and capturing the inherent relationship between factors. GAME-RL learns more effective evasion strategies, taking into account functionality preservation and the black-box setting. We conduct comprehensive experiments on various target models, demonstrating that GAME-RL significantly improves the evasion rate while maintaining acceptable levels of adversarial overhead.

Static analysis is a crucial protection layer that enables modern antivirus systems to address the rampant proliferation of malware. These systems are increasingly relying on deep neural networks (DNNs) to automatically extract reliable features and achieve outstanding detection accuracy. Since DNNs are known to be vulnerable to adversarial examples, several studies have proposed practical evasion attacks to generate adversarial perturbations that can evade malware detectors. These attacks, however, require specific designs for the given input sample, prohibiting them from large-scale deployment. Therefore, it is more practical to generate sample-agnostic perturbations that do not involve recalculations regardless of the input malware sample. To this end, we leverage an adversarial patch attack, which is a special type of adversarial attack that dose not know the sample being modified during the attack construction process. In particular, we propose a new adversarial attack against malware detection systems called MalPatch. It locates the nonfunctional part of malware for adversarial patch injection to protect its executability while generating adversarial examples based on different strategies. The generated patch can be injected into any malware sample, fooling the detector into classifying it as benign. Experimental results demonstrate that MalPatch is effective under different attack settings. In the white-box setting, MalPatch achieves 69%-78% success rates against DNN detectors based on raw byte features and 47%-96% success rates against four grayscale detectors based on image features. In the black-box setting, the success rates of MalPatch against the same models reach 54%-74% and 27%-42%, respectively. We conclude by discussing several of its potential countermeasures and the generality of our approach.

With the widespread application of machine learning techniques in malware detection, researchers have proposed various adversarial attack methods to generate adversarial examples (AEs) of malware, thereby evading detection. Previous studies have shown that the reinforcement learning (RL) framework can enable black-box attacks by performing a sequence of function-preserving operations, which produces functional evasive malware samples. However, it is difficult to obtain the useful guidance and feedbacks from the environment for agent training in the black-box scenario, which results in the RL framework being unable to learn the effective evasion policy. In this paper, we propose the Shapley prior and establish a prior-guidance-based RL framework, namely PSP-Mal, to generate AEs against Portable Executable (PE) malware detectors. Our framework improves on existing methods in three aspects: 1) We explore feature effects of the black-box model by computing Shapley values and further propose the Shapley prior to represent the expected impact of operations. 2) A novel prioritized experience utilization mechanism is established regarding the Shapley prior guidance in the RL framework. 3) The actions are expanded into item-content pairs and we use the Thompson sampling to choose effective content, which helps to reduce randomness and ensure repeatability. We compare the attack performance of our framework with other methods, and experimental results demonstrate that our algorithm is more effective. The evasion rates of PSP-Mal against the LightGBM models trained on EMBER and SOREL-20M reach 76.88% and 72.03%, respectively.

An adversary can fool deep neural network object detectors by generating adversarial noises. Most of the existing works focus on learning local visible noises in an adversarial "patch" fashion. However, the 2D patch attached to a 3D object tends to suffer from an inevitable reduction in attack performance as the viewpoint changes. To remedy this issue, this work proposes the Coated Adversarial Camouflage (CAC) to attack the detectors in arbitrary viewpoints. Unlike the patch trained in the 2D space, our camouflage generated by a conceptually different training framework consists of 3D rendering and dense proposals attack. Specifically, we make the camouflage perform 3D spatial transformations according to the pose changes of the object. Based on the multi-view rendering results, the top-n proposals of the region proposal network are fixed, and all the classifications in the fixed dense proposals are attacked simultaneously to output errors. In addition, we build a virtual 3D scene to fairly and reproducibly evaluate different attacks. Extensive experiments demonstrate the superiority of CAC over the existing attacks, and it shows impressive performance both in the virtual scene and the real world. This poses a potential threat to the security-critical computer vision systems.

Abstract Deep neural networks are vulnerable to adversarial examples. However, existing attacks exhibit relatively low efficacy in generating transferable adversarial examples. Improved transferability to address this issue is proposed via a rotation‐invariant attack method that maximizes the loss function w.r.t the random rotated image instead of the original input at each iteration, thus mitigating the high correlation between the adversarial examples and the source models and making the adversarial examples more transferable. Extensive experiments show that the proposed method can significantly improve the transferability of the adversarial examples with almost no extra computational cost and can be integrated into various methods. In addition, when this method is easily applied through a plug‐in, the average attack success rate against six robustly trained models increases by 5.4% over the state‐of‐the‐art baseline method, demonstrating its effectiveness and efficiency. The codes used are publicly available at https://github.com/YeXinD/Rotation‐Invariant‐Attack .

Attacker identification from network traffic is a common practice of cyberspace security management. However, network administrators cannot cover all security equipment due to the cyberspace management cost constraints, giving attackers the chance to escape from the surveillance of network security administrators by legitimate actions and to perform the attack in both physical domain and digital domain. Therefore, we proposed a hidden attack sequence detection method based on reinforcement learning to deal with the challenge through modeling the network administrators as an intelligent agent that learns their action policy from the interaction with the cyberspace environment. Following Deep Deterministic Policy Gradient (DDPG), the intelligent agent can not only discover the hidden attackers hiding in the legitimate action sequences but also reduce the cyberspace management cost. Furthermore, a dynamic reward DDPG method was proposed to improve defense performance, which set dynamic reward depending on the hidden attack sequences steps and agent’s check steps, compared to the fixed reward in common methods. Meanwhile, the method was verified in a simulated experimental cyberspace environment. Finally, the experimental results demonstrate that there are hidden attack sequences in cyberspace, and the proposed method can discover the hidden attack sequences. The dynamic reward DDPG shows superior performance in detecting hidden attackers, with a detection rate of 97.46%, which can improve the ability to discover hidden attackers and reduce the 6% cyberspace management cost compared to DDPG.

Averaging scheme has attracted extensive attention in deep learning as well as traditional machine learning. It achieves theoretically optimal convergence and also improves the empirical model performance. However, there is still a lack of sufficient convergence analysis for strongly convex optimization. Typically, the convergence about the last iterate of gradient descent methods, which is referred to as individual convergence, fails to attain its optimality due to the existence of logarithmic factor. In order to remove this factor, we first develop gradient descent averaging (GDA), which is a general projection-based dual averaging algorithm in the strongly convex setting. We further present primal-dual averaging for strongly convex cases (SC-PDA), where primal and dual averaging schemes are simultaneously utilized. We prove that GDA yields the optimal convergence rate in terms of output averaging, while SC-PDA derives the optimal individual convergence. Several experiments on SVMs and deep learning models validate the correctness of theoretical analysis and effectiveness of algorithms.

Searching for active modules, i.e., regions showing striking changes in molecular activity in biological networks is important to reveal regulatory and signaling mechanisms of biological systems. Most existing active modules identification methods are based on protein-protein interaction networks or metabolic networks, which require comprehensive and accurate prior knowledge. On the other hand, weighted gene co-expression networks (WGCNs) are purely constructed from gene expression profiles. However, existing WGCN analysis methods are designed for identifying functional modules but not capable of identifying active modules. There is an urgent need to develop an active module identification algorithm for WGCNs to discover regulatory and signaling mechanism associating with a given cellular response. To address this urgent need, we propose a novel algorithm called active modules on the multi-layer weighted (co-expression gene) network, based on a continuous optimization approach (AMOUNTAIN). The algorithm is capable of identifying active modules not only from single-layer WGCNs but also from multilayer WGCNs such as cross-species and dynamic WGCNs. We first validate AMOUNTAIN on a synthetic benchmark dataset. We then apply AMOUNTAIN to WGCNs constructed from Th17 differentiation gene expression datasets of human and mouse, which include a single layer, a cross-species two-layer and a multilayer dynamic WGCNs. The identified active modules from WGCNs are enriched by known protein-protein interactions, and more importantly, they reveal some interesting and important regulatory and signaling mechanisms of Th17 cell differentiation.

Background subtraction is a classic video processing task pervading in numerous visual applications such as video surveillance and traffic monitoring. Given the diversity and variability of real application scenes, an ideal background subtraction model should be robust to various scenarios. Even though deep-learning approaches have demonstrated unprecedented improvements, they often fail to generalize to unseen scenarios, thereby less suitable for extensive deployment. In this work, we propose to tackle cross-scene background subtraction via a two-phase framework that includes meta-knowledge learning and domain adaptation. Specifically, as we observe that meta-knowledge (i.e., scene-independent common knowledge) is the cornerstone for generalizing to unseen scenes, we draw on traditional frame differencing algorithms and design a deep difference network (DDN) to encode meta-knowledge especially temporal change knowledge from various cross-scene data (source domain) without intermittent foreground motion pattern. In addition, we explore a self-training domain adaptation strategy based on iterative evolution. With iteratively updated pseudo-labels, the DDN is continuously fine-tuned and evolves progressively toward unseen scenes (target domain) in an unsupervised fashion. Our framework could be easily deployed on unseen scenes without relying on their annotations. As evidenced by our experiments on the CDnet2014 dataset, it brings a significant improvement to background subtraction. Our method has a favorable processing speed (70 fps) and outperforms the best unsupervised algorithm and top supervised algorithm designed for unseen scenes by 9% and 3%, respectively.

Community structure, a foundational concept in understanding networks, is one of the most important properties of dynamic networks. A large number of dynamic community detection methods proposed are based on the temporal smoothness framework that the abrupt change of clustering within a short perio d is undesirable. However, how to improve the community detection performance by combining network topology information in a short period is a challenging problem. Additionally, previous efforts on utilizing such properties are insufficient. In this paper, we introduce the geometric structure of a network to represent the temporal smoothness in a short time and propose a novel Dynamic Graph Regularized Symmetric NMF method (DGR-SNMF) to detect the community in dynamic networks. This method combines geometric structure information sufficiently in current detecting process by Symmetric Non-negative Matrix Factorization (SNMF). We also prove the convergence of the iterative update rules by constructing auxiliary functions. Extensive experiments on multiple synthetic networks and two real-world datasets demonstrate that the proposed DGR-SNMF method outperforms the state-of-the-art algorithms on detecting dynamic community.

Since the number of malware is increasing rapidly, it continuously poses a risk to the field of network security. Attention mechanism has made great progress in the field of natural language processing. At the same time, there are many research studies based on malicious code API, which is also like semantic information. It is a worthy study to apply attention mechanism to API semantics. In this paper, we firstly study the characters of the API execution sequence and classify them into 17 categories. Secondly, we propose a novel feature extraction method based on API execution sequence according to its semantics and structure information. Thirdly, based on the API data characteristics and attention mechanism features, we construct a detection framework SLAM based on local attention mechanism and sliding window method. Experiments show that our model achieves a better performance, which is a higher accuracy of 0.9723.

Many well-known first-order gradient methods have been extended to cope with large-scale composite problems, which often arise as a regularized empirical risk minimization in machine learning. However, their optimal convergence is attained only in terms of the weighted average of past iterative solutions. How to make the individual convergence of stochastic gradient descent (SGD) optimal, especially for strongly convex problems has now become a challenging problem in the machine learning community. On the other hand, Nesterov's recent weighted averaging strategy succeeds in achieving the optimal individual convergence of dual averaging (DA) but it fails in the basic mirror descent (MD). In this paper, a new primal averaging (PA) gradient operation step is presented, in which the gradient evaluation is imposed on the weighted average of all past iterative solutions. We prove that simply modifying the gradient operation step in MD by PA strategy suffices to recover the optimal individual rate for general convex problems. Along this line, the optimal individual rate of convergence for strongly convex problems can also be achieved by imposing the strong convexity on the gradient operation step. Furthermore, we extend PA-MD to solve regularized nonsmooth learning problems in the stochastic setting, which reveals that PA strategy is a simple yet effective extra step toward the optimal individual convergence of SGD. Several real experiments on sparse learning and SVM problems verify the correctness of our theoretical analysis.

The extrapolation strategy raised by Nesterov, which can accelerate the convergence rate of gradient descent methods by orders of magnitude when dealing with smooth convex objective, has led to tremendous success in training machine learning tasks. In this article, the convergence of individual iterates of projected subgradient (PSG) methods for nonsmooth convex optimization problems is theoretically studied based on Nesterov's extrapolation, which we name individual convergence. We prove that Nesterov's extrapolation has the strength to make the individual convergence of PSG optimal for nonsmooth problems. In light of this consideration, a direct modification of the subgradient evaluation suffices to achieve optimal individual convergence for strongly convex problems, which can be regarded as making an interesting step toward the open question about stochastic gradient descent (SGD) posed by Shamir. Furthermore, we give an extension of the derived algorithms to solve regularized learning tasks with nonsmooth losses in stochastic settings. Compared with other state-of-the-art nonsmooth methods, the derived algorithms can serve as an alternative to the basic SGD especially in coping with machine learning problems, where an individual output is needed to guarantee the regularization structure while keeping an optimal rate of convergence. Typically, our method is applicable as an efficient tool for solving large-scale l1-regularized hinge-loss learning problems. Several comparison experiments demonstrate that our individual output not only achieves an optimal convergence rate but also guarantees better sparsity than the averaged solution.

Network representation learning aims at learning a low-dimensional vector for each node in a network, which has attracted increasing research interests recently. However, most existing approaches only use topology information of each node and ignore its attributes information. In this paper, we pro pose an Improved Attributed Node Random Walks(IANRW) framework, which constructs the neighborhood of an attributed node and then leverages the skip-gram model to perform node embeddings. The method can be able to flexibly incorporate both the topology and attribute information. Additionally, it can easily deal with missing data and be applied to large networks. Extensive experiments on six datasets show that IANRW outperforms many state-of-the-art embedding models and can improve various attributed networks mining tasks.

Role-based access control (RBAC) is widely adopted in network security management, and role mining technology has been extensively used to automatically generate user roles from datasets in a bottom-up way. However, almost all role mining methods discover the user roles from existing user-permission assignments, which neglect the dependency relationships between user permissions. To extend the ability of role mining technology, this paper proposes a novel role mining framework based on multi-domain information. The framework estimates the similarity between different permissions based on the fundamental information in the physical, network, and digital domains and attaches interdependent permissions to the same role. Three simulated network scenarios with different multi-domain configurations are used to validate the effectiveness of our method. The experimental results show that the method can not only capture the interdependent relationships between permissions, but also detect user roles and permissions more reasonably.

The explosive growth of malware variants poses a continuously and deeply evolving challenge to information security. Traditional malware detection methods require a lot of manpower. However, machine learning has played an important role on malware classification and detection, and it is easily spoofed by malware disguising to be benign software by employing self-protection techniques, which leads to poor performance for existing techniques based on the machine learning method. In this paper, we analyze the local maliciousness about malware and implement an anti-interference detection framework based on API fragments, which uses the LSTM model to classify API fragments and employs ensemble learning to determine the final result of the entire API sequence. We present our experimental results on Ali-Tianchi contest API databases. By comparing with the experiments of some common methods, it is proved that our method based on local maliciousness has better performance, which is a higher accuracy rate of 0.9734.

Setting up a multidimensional network is an important problem in complex networks and has become a future development trend in the fields of biological gene networks, social networks and so on. A multidimensional network comprises connections and attributes. Community detection in heterogeneous dat asets in different dimensions is more difficult than that in a single network. Traditional methods for dealing with multidimensional networks are ineffective, because of using supervised information or applying strategies for adjusting the graph structure of a single network. In this paper, we propose a semi-supervised community detection method for multidimensional heterogeneous networks. First, we generate a single network by integrating the multidimensional heterogeneous networks. The robust semi-supervised link adjustment strategy is then iteratively applied to the single network to make full use of dynamic supervised information for adding or removing links based on node entropy. Experimental results are obtained by five real multidimensional social datasets. The results show that the proposed method can effectively integrate heterogeneous data. The average accuracy rate and standard mutual information were 90.50% and 93.99%, respectively, representing improvements of 28.97% and 35.06%, respectively, over existing methods.

Machine-learning algorithms are widely applied in traffic classification and anomaly detection. Due to the tremendous traffic on the network, an extremely challenging question arises: how to efficiently and accurately detect the anomalous flow from the backbone network. One solution is proposed, online anomaly-detection scheme, which is based on the sparse feature selection method, Lasso. The sparse feature selection can be efficiently solved by reformulating the problem as an optimization problem with an ℓ1-ball constraint. At the evaluation stage, the authors preprocessed the raw data trace from the trans-Pacific backbone link between Japan and the United States and generated an evaluation data set. Their empirical study shows that the feature selection step can be solved quickly by applying the efficient Euclidean projection method; indeed, doing so resolves the feature selection step faster than using three classical ℓ1-min solvers. In terms of overall accuracy, true positive rate, false positive rate, precision, and F-measure, the proposed scheme improves the quality of detection. Copyright © 2015 John Wiley & Sons, Ltd.