Biblio

Learning-enabled components (LECs) are widely used in cyber-physical systems (CPS) since they can handle the uncertainty and variability of the environment and increase the level of autonomy. However, it has been shown that LECs such as deep neural networks (DNN) are not robust and adversarial examples can cause the model to make a false prediction. The paper considers the problem of efficiently detecting adversarial examples in LECs used for regression in CPS. The proposed approach is based on inductive conformal prediction and uses a regression model based on variational autoencoder. The architecture allows to take into consideration both the input and the neural network prediction for detecting adversarial, and more generally, out-of-distribution examples. We demonstrate the method using an advanced emergency braking system implemented in an open source simulator for self-driving cars where a DNN is used to estimate the distance to an obstacle. The simulation results show that the method can effectively detect adversarial examples with a short detection delay.

This research-to-practice work-in-progress addresses a new approach to cybersecurity education. The cyber security skills shortage is reaching prevalent proportions. The consensus in the STEM community is that the problem begins at k-12 schools with too few students interested in STEM subjects. One way to ensure a larger pipeline in cybersecurity is to train more high school teachers to not only teach cybersecurity in their schools or integrate cybersecurity concepts in their classrooms but also to promote IT security as an attractive career path. The proposed research will result in developing a unique and novel curriculum and scalable program in the area of cybersecurity and a set of powerful tools for a fun learning experience in cybersecurity education. In this project, we are focusing on the potential to advance research agendas in cybersecurity and train the future generation with cybersecurity skills and answer fundamental research questions that still exist in the blended learning methodologies for cybersecurity education and assessment. Leadership and entrepreneurship skills are also added to the mix to prepare students for real-world problems. Delivery methods, timing, format, pacing and outcomes alignment will all be assessed to provide a baseline for future research and additional synergy and integration with existing cybersecurity programs to expand or leverage for new cybersecurity and STEM educational research. This is a new model for cybersecurity education, leadership, and entrepreneurship and there is a possibility of a significant leap towards a more advanced cybersecurity educational methodology using this model. The project will also provide a prototype for innovation coupled with character-building and ethical leadership.

The time-varying properties of the wireless channel are a powerful source of information that can complement and enhance traditional security mechanisms. Therefore, we propose a cross-layer authentication mechanism that combines physical layer channel information and traditional authentication mechanism in LTE. To verify the feasibility of the proposed mechanism, we build a cross-layer authentication system that extracts the phase shift information of a typical UE and use the ensemble learning method to train the fingerprint map based on OAI LTE. Experimental results show that our cross-layer authentication mechanism can effectively prompt the security of LTE system.

Conventional adversarial defenses reduce classification accuracy whether or not a model is under attacks. Moreover, most of image processing based defenses are defeated due to the problem of obfuscated gradients. In this paper, we propose a new adversarial defense which is a defensive transform for both training and test images inspired by perceptual image encryption methods. The proposed method utilizes a block-wise pixel shuffling method with a secret key. The experiments are carried out on both adaptive and non-adaptive maximum-norm bounded white-box attacks while considering obfuscated gradients. The results show that the proposed defense achieves high accuracy (91.55%) on clean images and (89.66%) on adversarial examples with noise distance of 8/255 on CFAR-10 dataset. Thus, the proposed defense outperforms state-of-the-art adversarial defenses including latent adversarial training, adversarial training and thermometer encoding.

The security of deep learning becomes increasing important with the more and more related applications. The adversarial attack is the known method that makes the performance of deep learning network (DNN) decline rapidly. However, adversarial attack needs the gradient knowledge of the target networks to craft the specific adversarial examples, which is the white-box attack and hardly becomes true in reality. In this paper, we implement a black-box attack on DNN classifier via a functionally equivalent network without knowing the internal structure and parameters of the target networks. And we increase the entropy of the noise via deep convolution generative adversarial networks (DCGAN) to make it seems fuzzier, avoiding being probed and eliminated easily by adversarial training. Experiments show that this method can produce a large number of adversarial examples quickly in batch and the target network cannot improve its accuracy via adversarial training simply.

Discovering the potential vulnerabilities in software plays a crucial role in ensuring the security of computer system. This paper proposes a method that can assist security auditors with the analysis of source code. When security auditors identify new vulnerabilities, our method can be adopted to make a list of recommendations that may have the same vulnerabilities for the security auditors. Our method relies on graph representation to automatically extract the mode of PDG(program dependence graph, a structure composed of control dependence and data dependence). Besides, it can be applied to the vulnerability extrapolation scenario, thus reducing the amount of audit code. We worked on an open-source vulnerability test set called Juliet. According to the evaluation results, the clustering effect produced is satisfactory, so that the feature vectors extracted by the Graph2Vec model are applied to labeling and supervised learning indicators are adopted to assess the model for its ability to extract features. On a total of 12,000 small data sets, the training score of the model can reach up to 99.2%, and the test score can reach a maximum of 85.2%. Finally, the recommendation effect of our work is verified as satisfactory.

Ahstract-Vulnerability detection is an important topic of software engineering. To improve the effectiveness and efficiency of vulnerability detection, many traditional machine learning-based and deep learning-based vulnerability detection methods have been proposed. However, the impact of different factors on vulnerability detection is unknown. For example, classification models and vectorization methods can directly affect the detection results and code replacement can affect the features of vulnerability detection. We conduct a comparative study to evaluate the impact of different classification algorithms, vectorization methods and user-defined variables and functions name replacement. In this paper, we collected three different vulnerability code datasets. These datasets correspond to different types of vulnerabilities and have different proportions of source code. Besides, we extract and analyze the features of vulnerability code datasets to explain some experimental results. Our findings from the experimental results can be summarized as follows: (i) the performance of using deep learning is better than using traditional machine learning and BLSTM can achieve the best performance. (ii) CountVectorizer can improve the performance of traditional machine learning. (iii) Different vulnerability types and different code sources will generate different features. We use the Random Forest algorithm to generate the features of vulnerability code datasets. These generated features include system-related functions, syntax keywords, and user-defined names. (iv) Datasets without user-defined variables and functions name replacement will achieve better vulnerability detection results.

This work investigates the vulnerability of Gaussian Mixture Model (GMM) i-vector based speaker verification systems to adversarial attacks, and the transferability of adversarial samples crafted from GMM i-vector based systems to x-vector based systems. In detail, we formulate the GMM i-vector system as a scoring function of enrollment and testing utterance pairs. Then we leverage the fast gradient sign method (FGSM) to optimize testing utterances for adversarial samples generation. These adversarial samples are used to attack both GMM i-vector and x-vector systems. We measure the system vulnerability by the degradation of equal error rate and false acceptance rate. Experiment results show that GMM i-vector systems are seriously vulnerable to adversarial attacks, and the crafted adversarial samples are proved to be transferable and pose threats to neural network speaker embedding based systems (e.g. x-vector systems).

In recent times, an increasing amount of intelligent electronic devices (IEDs) are being deployed to make power systems more reliable and economical. While these technologies are necessary for realizing a cyber-physical infrastructure for future smart power grids, they also introduce new vulnerabilities in the grid to different cyber-attacks. Traditional methods such as state vector estimation (SVE) are not capable of identifying cyber-attacks while the geometric information is also injected as an attack vector. In this paper, a machine learning based smart grid attack identification method is proposed. The proposed method is carried out by first collecting smart grid power flow data for machine learning training purposes which is later used to classify the attacks. The performance of both the proposed SVM method and the traditional SVE method are validated on IEEE 14, 30, 39, 57 and 118 bus systems, and the performance regarding the scale of the power system is evaluated. The results show that the SVM-based method performs better than the SVE-based in attack identification over a much wider scale of power systems.

Security is always a main concern in every domain, due to a rise in crime rate in a crowded event or suspicious lonely areas. Abnormal detection and monitoring have major applications of computer vision to tackle various problems. Due to growing demand in the protection of safety, security and personal properties, needs and deployment of video surveillance systems can recognize and interpret the scene and anomaly events play a vital role in intelligence monitoring. This paper implements automatic gun (or) weapon detection using a convolution neural network (CNN) based SSD and Faster RCNN algorithms. Proposed implementation uses two types of datasets. One dataset, which had pre-labelled images and the other one is a set of images, which were labelled manually. Results are tabulated, both algorithms achieve good accuracy, but their application in real situations can be based on the trade-off between speed and accuracy.

In recent years, with the continuous development of various new Internet technologies, big data, cloud computing and other technologies have been widely used in work and life. The further improvement of data scale and computing capability has promoted the breakthrough development of artificial intelligence technology. The generalization and classification of financial science and technology not only have a certain impact on the traditional financial business, but also put forward higher requirements for commercial banks to operate financial science and technology business. Artificial intelligence brings fresh experience to financial services and is conducive to increasing customer stickiness. Artificial intelligence technology helps the standardization, modeling and intelligence of banking business, and helps credit decision-making, risk early warning and supervision. This paper first discusses the influence of artificial intelligence on financial innovation, and on this basis puts forward measures for the innovation and development of bank financial science and technology. Finally, it discusses the problem of computer information security management in bank financial innovation in the era of artificial intelligence.

Threat Actor Attribution is the task of identifying an attacker responsible for an attack. This often requires expert analysis and involves a lot of time. There had been attempts to detect a threat actor using machine learning techniques that use information obtained from the analysis of malware samples. These techniques will only be able to identify the attack, and it is trivial to guess the attacker because various attackers may adopt an attack method. A state-of-the-art method performs attribution of threat actors from text reports using Machine Learning and NLP techniques using Threat Intelligence reports. We use the same set of Threat Reports of Advanced Persistent Threats (APT). In this paper, we propose a Deep Learning architecture to attribute Threat actors based on threat reports obtained from various Threat Intelligence sources. Our work uses Neural Networks to perform the task of attribution and show that our method makes the attribution more accurate than other techniques and state-of-the-art methods.

Nowadays, the networks attacks became very sophisticated and hard to be recognized, The traditional types of intrusion detection systems became inefficient in predicting new types of attacks. As the IDS is an important factor in securing the network in the real time, many new effective IDS approaches have been proposed. In this paper, we intend to discuss different Artificial Neural Networks based IDS approaches, also we are going to categorize them in four categories (normal ANN, DNN, CNN, RNN) and make a comparison between them depending on different performance parameters (accuracy, FNR, FPR, training time, epochs and the learning rate) and other factors like the network structure, the classification type, the used dataset. At the end of the survey, we will mention the merits and demerits of each approach and suggest some enhancements to avoid the noticed drawbacks.

Anomaly detection classification technology based on deep learning is one of the crucial technologies supporting network security. However, as the data increasing, this traditional model cannot guarantee that the false alarm rate is minimized while meeting the high detection rate. Additionally, distribution of imbalanced abnormal samples will lead to an increase in the error rate of the classification results. In this work, since CNN is effective in network intrusion classification, we embed a compressed feature layer in CNN (Convolutional Neural Networks). The purpose is to improve the efficiency of network intrusion detection. After our model was trained for 55 epochs and we set the learning rate of the model to 0.01, the detection rate reaches over 98%.

The new switching architecture of the neural networks was proposed. The switching neural networks consist of the neurons and the switchers. The goal is to reduce expenses on the artificial neural network design and training. For realization of complex models, algorithms and methods of management the neural networks of the big size are required. The number of the interconnection links “everyone with everyone” grows with the number of neurons. The training of big neural networks requires the resources of supercomputers. Time of training of neural networks also depends on the number of neurons in the network. Switching neural networks are divided into fragments connected by the switchers. Training of switcher neuron network is provided by fragments. On the basis of switching neural networks the devices of associative memory were designed with the number of neurons comparable to the human brain.

The object of research is information security. The tools used for research are artificial neural networks. The goal is to increase the cryptography security. The problems are: the big volume of information, the expenses for neural networks design and training. It is offered to use the neural network for the cryptographic transformation of information.

Enforcing security and resilience in a cloud platform is an essential but challenging problem due to the presence of a large number of heterogeneous applications running on shared resources. A security analysis system that can detect threats or malware must exist inside the cloud infrastructure. Much research has been done on machine learning-driven malware analysis, but it is limited in computational complexity and detection accuracy. To overcome these drawbacks, we proposed a new malware detection system based on the concept of clustering and trend micro locality sensitive hashing (TLSH). We used Cuckoo sandbox, which provides dynamic analysis reports of files by executing them in an isolated environment. We used a novel feature extraction algorithm to extract essential features from the malware reports obtained from the Cuckoo sandbox. Further, the most important features are selected using principal component analysis (PCA), random forest, and Chi-square feature selection methods. Subsequently, the experimental results are obtained for clustering and non-clustering approaches on three classifiers, including Decision Tree, Random Forest, and Logistic Regression. The model performance shows better classification accuracy and false positive rate (FPR) as compared to the state-of-the-art works and non-clustering approach at significantly lesser computation cost.

Whenever any internet user visits a website, a scripting language runs in the background known as JavaScript. The embedding of malicious activities within the script poses a great threat to the cyberworld. Attackers take advantage of the dynamic nature of the JavaScript and embed malicious code within the website to download malware and damage the host. JavaScript developers obfuscate the script to keep it shielded from getting detected by the malware detectors. In this paper, we propose a novel technique for analysing and detecting JavaScript using sandbox assisted ensemble model. We extract the payload using malware-jail sandbox to get the real script. Upon getting the extracted script, we analyse it to define the features that are needed for creating the dataset. We compute Pearson's r between every feature for feature extraction. An ensemble model consisting of Sequential Minimal Optimization (SMO), Voted Perceptron and AdaBoost algorithm is used with voting technique to detect malicious JavaScript. Experimental results show that our proposed model can detect obfuscated and de-obfuscated malicious JavaScript with an accuracy of 99.6% and 0.03s detection time. Our model performs better than other state-of-the-art models in terms of accuracy and least training and detection time.

Learning-based approaches often outperform hand-coded algorithmic solutions for many problems in robotics. However, learning long-horizon tasks on real robot hardware can be intractable, and transferring a learned policy from simulation to reality is still extremely challenging. We present a novel approach to model-free reinforcement learning that can leverage existing sub-optimal solutions as an algorithmic prior during training and deployment. During training, our gated fusion approach enables the prior to guide the initial stages of exploration, increasing sample-efficiency and enabling learning from sparse long-horizon reward signals. Importantly, the policy can learn to improve beyond the performance of the sub-optimal prior since the prior's influence is annealed gradually. During deployment, the policy's uncertainty provides a reliable strategy for transferring a simulation-trained policy to the real world by falling back to the prior controller in uncertain states. We show the efficacy of our Multiplicative Controller Fusion approach on the task of robot navigation and demonstrate safe transfer from simulation to the real world without any fine-tuning. The code for this project is made publicly available at https://sites.google.com/view/mcf-nav/home.

With its huge real-world demands, large-scale confidential computing still cannot be supported by today's Trusted Execution Environment (TEE), due to the lack of scalable and effective protection of high-throughput accelerators like GPUs, FPGAs, and TPUs etc. Although attempts have been made recently to extend the CPU-like enclave to GPUs, these solutions require change to the CPU or GPU chips, may introduce new security risks due to the side-channel leaks in CPU-GPU communication and are still under the resource constraint of today's CPU TEE.To address these problems, we present the first Heterogeneous TEE design that can truly support large-scale compute or data intensive (CDI) computing, without any chip-level change. Our approach, called HETEE, is a device for centralized management of all computing units (e.g., GPUs and other accelerators) of a server rack. It is uniquely designed to work with today's data centres and clouds, leveraging modern resource pooling technologies to dynamically compartmentalize computing tasks, and enforce strong isolation and reduce TCB through hardware support. More specifically, HETEE utilizes the PCIe ExpressFabric to allocate its accelerators to the server node on the same rack for a non-sensitive CDI task, and move them back into a secure enclave in response to the demand for confidential computing. Our design runs a thin TCB stack for security management on a security controller (SC), while leaving a large set of software (e.g., AI runtime, GPU driver, etc.) to the integrated microservers that operate enclaves. An enclaves is physically isolated from others through hardware and verified by the SC at its inception. Its microserver and computing units are restored to a secure state upon termination.We implemented HETEE on a real hardware system, and evaluated it with popular neural network inference and training tasks. Our evaluations show that HETEE can easily support the CDI tasks on the real-world scale and incurred a maximal throughput overhead of 2.17% for inference and 0.95% for training on ResNet152.

Spiking Neural Networks (SNNs) claim to present many advantages in terms of biological plausibility and energy efficiency compared to standard Deep Neural Networks (DNNs). Recent works have shown that DNNs are vulnerable to adversarial attacks, i.e., small perturbations added to the input data can lead to targeted or random misclassifications. In this paper, we aim at investigating the key research question: "Are SNNs secure?" Towards this, we perform a comparative study of the security vulnerabilities in SNNs and DNNs w.r.t. the adversarial noise. Afterwards, we propose a novel black-box attack methodology, i.e., without the knowledge of the internal structure of the SNN, which employs a greedy heuristic to automatically generate imperceptible and robust adversarial examples (i.e., attack images) for the given SNN. We perform an in-depth evaluation for a Spiking Deep Belief Network (SDBN) and a DNN having the same number of layers and neurons (to obtain a fair comparison), in order to study the efficiency of our methodology and to understand the differences between SNNs and DNNs w.r.t. the adversarial examples. Our work opens new avenues of research towards the robustness of the SNNs, considering their similarities to the human brain's functionality.

In the application scenarios of cloud computing, big data, and mobile Internet, covert and diverse network attacks have become a serious problem that threatens the security of enterprises and personal information assets. Abnormal network behaviour detection based on network behaviour characteristics has become an important means to protect network security. However, existing frameworks do not make full use of the characteristics of the correlation between continuous network behaviours, and do not use an algorithm that can process time-series data or process the original feature set into time-series data to match the algorithm. This paper proposes a time-series abnormal network behaviour detection framework. The framework consists of two parts: an algorithm model (DBN-BiGRU) that combines Deep Belief Network (DBN) and Bidirectional Gated Recurrent Unit (BiGRU), and a pre-processing scheme that processes the original feature analysis files of CICIDS2017 to good time-series data. This detection framework uses past and future behaviour information to determine current behaviours, which can improve accuracy, and can adapt to the large amount of existing network traffic and high-dimensional characteristics. Finally, this paper completes the training of the algorithm model and gets the test results. Experimental results show that the prediction accuracy of this framework is as high as 99.82%, which is better than the traditional frameworks that do not use time-series information.

This letter presents a novel median filtering forensics approach, based on a convolutional neural network (CNN) with an adaptive filtering layer (AFL), which is built in the discrete cosine transform (DCT) domain. Using the proposed AFL, the CNN can determine the main frequency range closely related with the operational traces. Then, to automatically learn the multi-scale manipulation features, a multi-scale convolutional block is developed, exploring a new multi-scale feature fusion strategy based on the maxout function. The resultant features are further processed by a convolutional stream with pooling and batch normalization operations, and finally fed into the classification layer with the Softmax function. Experimental results show that our proposed approach is able to accurately detect the median filtering manipulation and outperforms the state-of-the-art schemes, especially in the scenarios of low image resolution and serious compression loss.

In this paper, we introduce a new digital image forensics approach called forensic similarity, which determines whether two image patches contain the same forensic trace or different forensic traces. One benefit of this approach is that prior knowledge, e.g., training samples, of a forensic trace is not required to make a forensic similarity decision on it in the future. To do this, we propose a two-part deep-learning system composed of a convolutional neural network-based feature extractor and a three-layer neural network, called the similarity network. This system maps the pairs of image patches to a score indicating whether they contain the same or different forensic traces. We evaluated the system accuracy of determining whether two image patches were captured by the same or different camera model and manipulated by the same or a different editing operation and the same or a different manipulation parameter, given a particular editing operation. Experiments demonstrate applicability to a variety of forensic traces and importantly show efficacy on “unknown” forensic traces that were not used to train the system. Experiments also show that the proposed system significantly improves upon prior art, reducing error rates by more than half. Furthermore, we demonstrated the utility of the forensic similarity approach in two practical applications: forgery detection and localization, and database consistency verification.

Insider threat makes great damage to the security of information system, traditional security methods are extremely difficult to work. Insider attack identification plays an important role in insider threat detection. Monitoring user's abnormal behavior is an effective method to detect impersonation, this method is applied to insider threat identification, to built user's behavior attribute information database based on weights changeable feedback tree augmented Bayes network, but data is massive, using the dimensionality reduction based on rough set, to establish the process information model of user's behavior attribute. Using the minimum risk Bayes decision can effectively identify the real identity of the user when user's behavior departs from the characteristic model.