Search

Illicit cryptocurrency mining has become one of the prevalent methods for monetization of computer security incidents. In this attack, victims’ computing resources are abused to mine cryptocurrency for the benefit of attackers. The most popular illicitly mined digital coin is Monero as it provides strong anonymity and is efficiently mined on CPUs.Illicit mining crucially relies on communication between compromised systems and remote mining pools using the de facto standard protocol Stratum. While prior research primarily focused on endpoint-based detection of in-browser mining, in this paper, we address network-based detection of cryptomining malware in general. We propose XMR-Ray, a machine learning detector using novel features based on reconstructing the Stratum protocol from raw NetFlow records. Our detector is trained offline using only mining traffic and does not require privacy-sensitive normal network traffic, which facilitates its adoption and integration.In our experiments, XMR-Ray attained 98.94% detection rate at 0.05% false alarm rate, outperforming the closest competitor. Our evaluation furthermore demonstrates that it reliably detects previously unseen mining pools, is robust against common obfuscation techniques such as encryption and proxies, and is applicable to mining in the browser or by compiled binaries. Finally, by deploying our detector in a large university network, we show its effectiveness in protecting real-world systems.

In security-sensitive applications, the success of machine learning depends on a thorough vetting of their resistance to adversarial data. In one pertinent, well-motivated attack scenario, an adversary may attempt to evade a deployed system at test time by carefully manipulating attack samples. In this work, we present a simple but effective gradient-based approach that can be exploited to systematically assess the security of several, widely-used classification algorithms against evasion attacks. Following a recently proposed framework for security evaluation, we simulate attack scenarios that exhibit different risk levels for the classifier by increasing the attacker's knowledge of the system and her ability to manipulate attack samples. This gives the classifier designer a better picture of the classifier performance under evasion attacks, and allows him to perform a more informed model selection (or parameter setting). We evaluate our approach on the relevant security task of malware detection in PDF files, and show that such systems can be easily evaded. We also sketch some countermeasures suggested by our analysis., Comment: In this paper, in 2013, we were the first to introduce the notion of evasion attacks (adversarial examples) created with high confidence (instead of minimum-distance misclassifications), and the notion of surrogate learners (substitute models). These two concepts are now widely re-used in developing attacks against deep networks (even if not always referring to the ideas reported in this work). arXiv admin note: text overlap with arXiv:1401.7727

Whenever machine learning is used to prevent illegal or unsanctioned activity and there is an economic incentive, adversaries will attempt to circumvent the protection provided. Constraints on how adversaries can manipulate training and test data for classifiers used to detect suspicious behavior make problems in this area tractable and interesting. This special issue highlights papers that span many disciplines including email spam detection, computer intrusion detection, and detection of web pages deliberately designed to manipulate the priorities of pages returned by modern search engines. The four papers in this special issue provide a standard taxonomy of the types of attacks that can be expected in an adversarial framework, demonstrate how to design classifiers that are robust to deleted or corrupted features, demonstrate the ability of modern polymorphic engines to rewrite malware so it evades detection by current intrusion detection and antivirus systems, and provide approaches to detect web pages designed to manipulate web page scores returned by search engines. We hope that these papers and this special issue encourages the multidisciplinary cooperation required to address many interesting problems in this relatively new area including predicting the future of the arms races created by adversarial learning, developing effective long-term defensive strategies, and creating algorithms that can process the massive amounts of training and test data available for internet-scale problems.

The article presents a general view of a class of decomposition algorithms for training Support Vector Machines (SVM) which are motivated by the method of feasible directions. The first such algorithm for the pattern recognition SVM has been proposed in Joachims, T. (1999, Scholkopf et al. (Eds.) Advances in kernel methods-Support vector learning (pp. 185–208). MIT Press). Its extension to the regression SVM—the maximal inconsistency algorithm—has been recently presented by the author (Laskov, 2000, Solla, Leen, & Muller (Eds.) Advances in neural information processing systems 12 (pp. 484–490). MIT Press). A detailed account of both algorithms is carried out, complemented by theoretical investigation of the relationship between the two algorithms. It is proved that the two algorithms are equivalent for the pattern recognition SVM, and the feasible direction interpretation of the maximal inconsistency algorithm is given for the regression SVM. The experimental results demonstrate an order of magnitude decrease of training time in comparison with training without decomposition, and, most importantly, provide experimental evidence of the linear convergence rate of the feasible direction decomposition algorithms.

Learning-based classifiers are increasingly used for detection of various forms of malicious data. However, if they are deployed online, an attacker may attempt to evade them by manipulating the data. Examples of such attacks have been previously studied under the assumption that an attacker has full knowledge about the deployed classifier. In practice, such assumptions rarely hold, especially for systems deployed online. A significant amount of information about a deployed classifier system can be obtained from various sources. In this paper, we experimentally investigate the effectiveness of classifier evasion using a real, deployed system, PDFrate, as a test case. We develop a taxonomy for practical evasion strategies and adapt known evasion algorithms to implement specific scenarios in our taxonomy. Our experimental results reveal a substantial drop of PDFrate's classification scores and detection accuracy after it is exposed even to simple attacks. We further study potential defense mechanisms against classifier evasion. Our experiments reveal that the original technique proposed for PDFrate is only effective if the executed attack exactly matches the anticipated one. In the discussion of the findings of our study, we analyze some potential techniques for increasing robustness of learning-based systems against adversarial manipulation of data.

In this chapter we introduce basic concepts and ideas of the Support Vector Machines (SVM). In the first section we formulate the learning problem in a statistical framework. A special focus is put on the concept of consistency, which leads to the principle of structural risk minimization (SRM). Application of these ideas to classification problems brings us to the basic, linear formulation of the SVM, described in Sect. 30.3. We then introduce the so called “kernel trick” as a tool for building a non-linear SVM as well as applying an SVM to non-vectorial data (Sect. 30.4). The practical issues of implementation of the SVM training algorithms and the related optimization problems are the topic of Sect. 30.5. Extensions of the SVM algorithms for the problems of non-linear regression and novelty detection are presented in Sect. 30.6. A brief description of the most successful applications of the SVM is given in Sect. 30.7. Finally, in the last Sect. 30.8 we summarize the main ideas of the chapter.

Despite the recent security improvements in Adobe's PDF viewer, its underlying code base remains vulnerable to novel exploits. A steady flow of rapidly evolving PDF malware observed in the wild substantiates the need for novel protection instruments beyond the classical signature-based scanners. In this contribution we present a technique for detection of JavaScript-bearing malicious PDF documents based on static analysis of extracted JavaScript code. Compared to previous work, mostly based on dynamic analysis, our method incurs an order of magnitude lower run-time overhead and does not require special instrumentation. Due to its efficiency we were able to evaluate it on an extremely large real-life dataset obtained from the VirusTotal malware upload portal. Our method has proved to be effective against both known and unknown malware and suitable for large-scale batch processing.

Learning for security applications is an emerging field where adaptive approaches are needed but are complicated by changing adversarial behavior. Traditional approaches to learning assume benign errors in data and thus may be vulnerable to adversarial errors. In this paper, we incorporate the notion of adversarial corruption directly into the learning framework and derive a new criteria for classifier robustness to adversarial contamination.

Hosts infected with malicious software, so called malware, are ubiquitous in today's computer networks. The means whereby malware can infiltrate a network are manifold and range from exploiting of software vulnerabilities to tricking a user into executing malicious code. Monitoring and detection of all possible infection vectors is intractable in practice. Hence, we approach the problem of detecting malicious software at a later point when it initiates contact with its maintainer; a process referred to as "phoning home". In particular, we introduce Botzilla, a method for detection of malware communication, which proceeds by repetitively recording network traffic of malware in a controlled environment and generating network signatures from invariant content patterns. Experiments conducted at a large university network demonstrate the ability of Botzilla to accurately identify malware communication in network traffic with very low false-positive rates.

The growing amount of web-based attacks poses a severe threat to the security of web applications. Signature-based detection techniques increasingly fail to cope with the variety and complexity of novel attack instances. As a remedy, we introduce a protocol-aware reverse HTTP proxy TokDoc (the token doctor), which intercepts requests and decides on a per-token basis whether a token requires automatic "healing". In particular, we propose an intelligent mangling technique, which, based on the decision of previously trained anomaly detectors, replaces suspicious parts in requests by benign data the system has seen in the past. Evaluation of our system in terms of accuracy is performed on two real-world data sets and a large variety of recent attacks. In comparison to state-of-the-art anomaly detectors, TokDoc is not only capable of detecting most attacks, but also significantly outperforms the other methods in terms of false positives. Runtime measurements show that our implementation can be deployed as an inline intrusion prevention system.

We propose a framework for quantitative security analysis of machine learning methods. The key parts of this framework are the formal specification of a deployed learning model and attacker's constraints, the computation of an optimal attack, and the derivation of an upper bound on adversarial impact. We exemplarily apply the framework for the analysis of one specific learning scenario, online centroid anomaly detection, and experimentally verify the tightness of obtained theoretical bounds.

New overlay-based services aim to provide properties like resilience, availability or QoS. To achieve this, automatic organization and optimization is required, which again demands accurate information on the network. Collecting and exchanging this data has a significant impact on the network, especially if several overlays are used on the same host. In this paper, we provide a survey of the current state of the art and identify challenges which must be addressed in order for new overlay-based services to be successful. We present our own solution CLIO, a cross-layer information service for overlays. CLIO provides information for the automatic creation and optimization of overlays. The service supports multiple overlays on the same node, the provided information is overlay-independent, and collected information is reused.

Malicious software in form of Internet worms, computer viruses, and Trojan horses poses a major threat to the security of networked systems. The diversity and amount of its variants severely undermine the effectiveness of classical signature-based detection. Yet variants of malware families share typical behavioral patternsreflecting its origin and purpose. We aim to exploit these shared patterns for classification of malware and propose a method for learning and discrimination of malware behavior. Our method proceeds in three stages: (a) behavior of collected malware is monitored in a sandbox environment, (b) based on a corpus of malware labeled by an anti-virus scanner a malware behavior classifieris trained using learning techniques and (c) discriminative features of the behavior models are ranked for explanation of classification decisions. Experiments with different heterogeneous test data collected over several months using honeypots demonstrate the effectiveness of our method, especially in detecting novelinstances of malware families previously not recognized by commercial anti-virus software.

A frequent problem in anomaly detection is to decide among different feature sets to be used. For example, various features are known in network intrusion detection based on packet headers, content byte streams or application level protocol parsing. A method for automatic feature selection in anomaly detection is proposed which determines optimal mixture coeffcients for various sets of features. The method generalizes the support vector data description (SVDD) and can be expressed as a semi-innite linear program that can be solved with standard techniques. The case of a single feature set can be handled as a particular case of the proposed method. The experimental evaluation of the new method on unsanitized HTTP data demonstrates that detectors using automatically selected features attain competitive performance, while sparing practitioners from a priori decisions on feature sets to be used. A frequent problem in anomaly detection is to decide among different feature sets to be used. For example, various features are known in network intrusion detection based on packet headers, content byte streams or application level protocol parsing. A method for automatic feature selection in anomaly detection is proposed which determines optimal mixture coeffcients for various sets of features. The method generalizes the support vector data description (SVDD) and can be expressed as a semi-innite linear program that can be solved with standard techniques. The case of a single feature set can be handled as a particular case of the proposed method. The experimental evaluation of the new method on unsanitized HTTP data demonstrates that detectors using automatically selected features attain competitive performance, while sparing practitioners from a priori decisions on feature sets to be used.

Current Voice-over-IP infrastructures lack defenses against unexpected network threats, such as zero-day exploits and computer worms. The possibility of such threats originates from the ongoing convergence of telecommunication and IP network infrastructures. As a countermeasure, we propose a self-learning system for detection of unknown and novel attacks in the Session Initiation Protocol (SIP). The system identifies anomalous content by embedding SIP messages to a feature space and determining deviation from a model of normality. The system adapts to network changes by automatically retraining itself while being hardened against targeted manipulations. Experiments conducted with realistic SIP traffic demonstrate the high detection performance of the proposed system at low false-positive rates.

In this paper, we propose a method for network intrusion detection based on language models. Our method proceeds by extracting language features such as n-grams and words from connection payloads and applying unsupervised anomaly detection—without prior learning phase or presence of labeled data. The essential part of this procedure is linear-time computation of similarity measures between language models of connection payloads. Particular patterns in these models decisive for differentiation of attacks and normal data can be traced back to attack semantics and utilized for automatic generation of attack signatures. Results of experiments conducted on two datasets of network traffic demonstrate the importance of high-order n-grams and variable-length language models for detection of unknown network attacks. An implementation of our system achieved detection accuracy of over 80% with no false positives on instances of recent remote-to-local attacks in HTTP, FTP and SMTP traffic.

Application and development of specialized machine learning techniques is gaining increasing attention in the intrusion detection community. A variety of learning techniques proposed for different intrusion detection problems can be roughly classified into two broad categories: supervised (classification) and unsupervised (anomaly detection and clustering). In this contribution we develop an experimental framework for comparative analysis of both kinds of learning techniques. In our framework we cast unsupervised techniques into a special case of classification, for which training and model selection can be performed by means of ROC analysis. We then investigate both kinds of learning techniques with respect to their detection accuracy and ability to detect unknown attacks.

Fingerspelling recognition and handshape recognition are two examples of real-world, multi-class recognition problems consisting of 26 and 78 classes respectively. While it is theoretically possible to solve any multi-class problem with a single "smart" classifier the complexity of such a classifier is usually prohibitively high. This paper looks at several approaches to solving a numerous multi-class recognition problem and discusses in detail a method involving coded output. Experiments are conducted using biomechanical data from a human hand as input, but work is continuing concerning the extraction of this data from multi-view hand images alone. Code generation is discussed and results are presented for several different coded output cases including the Hamming, Golay, and several hybrid codes. Conclusions show that the recognition accuracy increases proportionally to code length.

Recovery of motion parameters and point correspondences is a fundamental problem in computer vision. Although a great deal of research has been done solving rigid motion, nonrigid motion analysis has only recently been addressed, and is gaining interest due to its wide range of applications. This thesis introduces a novel method for estimating motion parameters and point correspondences between 2D curves and 3D surfaces under small nonrigid deformations. We use curvature to track point correspondences and motion parameters between 2D curves under small deformations. Curvature is invariant to the type of parameterization used, and to rigid transformations. These properties help us to deal with purely nonrigid motion, as curvature is not affected by translation/rotation of a given point. The motion parameter in this analysis is a point function, called the displacement function. In this work, it was defined as a polynomial function of a point on the curve under motion. Simulations are performed for illustrating accuracy of the algorithm's estimation by generating deformations on different curves. Real data experiments are also performed, using MR grid data, breast contours from mammogram images and contours of the left ventricle of heart. Three approaches are presented in this thesis for tracking point correspondences, and estimating motion parameters in 3D nonrigid motion. Unlike the 2D case, the motion parameter used here is a vector point function (not scalar), also called displacement function. It it represents both direction and magnitude of deformation at a given point. The first approach involves the use of changes in the "discriminant" for nonrigid motion analysis. Discriminant is defined as the infinitesimal area of surface at a given point. The second approach involves the use of changes in the unit normal at any given point on a surface under deformation. The third approach uses the challenges in Gaussian curvature along with the changes in unit normal direction. In all these approaches, the displacement function can be any point function representing the deformation involved. In our work, we have chosen a polynomial vector function to approximate the deformation. Simulations are performed for each approach by generating nonrigid motion on an ellipsoidal data set to illustrate performance and accuracy of the derived algorithms. Then, the algorithm is tested on a sequence of facial range images. A total of 16 sets of images were used in this experiment, where eight persons have generated two expressions each. We demonstrate the correct correspondence recovery by tracking points on the face during each facial expression. The correspondences estimation is also evaluated by comparing the algorithm's output against manual tracking of feature points by different users. In addition, nonrigid motion segmentation is successfully performed on these images. The motion parameters generated by the algorithm have also been used to demonstrate the abnormality in cardiac motion from CT images.

The paper presents two useful extensions of the incremental SVM in the context of online learning. An online support vector data description algorithm enables application of the online paradigm to unsupervised learning. Furthermore, online learning can be used in the large-scale classification problems to limit the memory requirements for storage of the kernel matrix. The proposed algorithms are evaluated on the task of online monitoring of EEG data, and on the classification task of learning the USPS dataset with a-priori chosen working set size.

We explore recognition implications of understanding gesture communication, having chosen American sign language as an example of a gesture language. An instrumented glove and specially developed software have been used for data collection and labeling. We address the problem of recognizing dynamic signing, i.e. signing performed at natural speed. Two neural network architectures have been used for recognition of different types of finger-spelled sentences. Experimental results are presented suggesting that two features of signing affect recognition accuracy: signing frequency which to a large extent can be accounted for by training a network on the samples of the respective frequency; and coarticulation effect which a network fails to identify. As a possible solution to coarticulation problem two post-processing algorithms for temporal segmentation are proposed and experimentally evaluated.

Background Bioinformatics data analysis often deals with additive mixtures of signals for which only class labels are known. Then, the overall goal is to estimate class related signals for data mining purposes. A convenient application is metabolic monitoring of patients using infrared spectroscopy. Within an infrared spectrum each single compound contributes quantitatively to the measurement. Results In this work, we propose a novel factorization technique for additive signal factorization that allows learning from classified samples. We define a composed loss function for this task and analytically derive a closed form equation such that training a model reduces to searching for an optimal threshold vector. Our experiments, carried out on synthetic and clinical data, show a sensitivity of up to 0.958 and specificity of up to 0.841 for a 15-class problem of disease classification. Using class and regression information in parallel, our algorithm outperforms linear SVM for training cases having many classes and few data. Conclusions The presented factorization method provides a simple and generative model and, therefore, represents a first step towards predictive factorization methods.