Search

Owing to increasingly important drivers for change, such as automation, digitisation, and dynamic demand patterns, simulation models of production systems become outdated rapidly. At the same time, building simulation models often requires much time, cost, and expertise, especially when dealing with complex job shop production systems. To address these challenges, an automated simulation model generation (ASMG) framework for material flow simulation of production systems is presented. This framework contains multiple approaches to infer routeing, control and temporal aspects from event-based data. To achieve this, methods from process mining (PM) and machine learning (ML) are applied. Additionally, the suitability of Coloured Petri Nets (CPNs) to serve as conceptual and operational simulation models is examined. The inferred simulation models have high validity when compared to the real system concerning the KPIs machine utilisation, throughput, and work in process. It is shown, that most model elements can be inferred very well, in particular process routes, processing times, and resource selection rules. This proof of concept serves as a foundation for research on detection approaches inferring further model elements such as setup times accurately. [ABSTRACT FROM AUTHOR]

Artificial intelligence (AI) is being ubiquitously adopted to automate processes in science and industry. However, due to its often intricate and opaque nature, AI has been shown to possess inherent vulnerabilities which can be maliciously exploited with adversarial AI, potentially putting AI users and developers at both cyber and physical risk. In addition, there is insufficient comprehension of the real-world effects of adversarial AI and an inadequacy of AI security examinations; therefore, the growing threat landscape is unknown for many AI solutions. To mitigate this issue, we propose one of the first red team frameworks for evaluating the AI security of maritime autonomous systems. The framework provides operators with a proactive (secure by design) and reactive (post-deployment evaluation) response to securing AI technology today and in the future. This framework is a multi-part checklist, which can be tailored to different systems and requirements. We demonstrate this framework to be highly effective for a red team to use to uncover numerous vulnerabilities within a real-world maritime autonomous systems AI, ranging from poisoning to adversarial patch attacks. The lessons learned from systematic AI red teaming can help prevent MAS-related catastrophic events in a world with increasing uptake and reliance on mission-critical AI. [ABSTRACT FROM AUTHOR]

Human–robot interaction (HRI) in human social environments (HSEs) poses unique challenges for robot perception systems, which must combine asynchronous, heterogeneous data streams in real time. Multimodal perception systems are well-suited for HRI in HSEs and can provide more rich, robust interaction for robots operating among humans. In this article, we provide an overview of multimodal perception systems being used in HSEs, which is intended to be an introduction to the topic and summary of relevant trends, techniques, resources, challenges, and terminology. We surveyed 15 peer-reviewed robotics and HRI publications over the past 10+ years, providing details about the data acquisition, processing, and fusion techniques used in 65 multimodal perception systems across various HRI domains. Our survey provides information about hardware, software, datasets, and methods currently available for HRI perception research, as well as how these perception systems are being applied in HSEs. Based on the survey, we summarize trends, challenges, and limitations of multimodal human perception systems for robots, then identify resources for researchers and developers and propose future research areas to advance the field. [ABSTRACT FROM AUTHOR]

Electrical Impedance Tomography is a noninvasive imaging technique that employs electrical currents to reveal the internal conductivity distribution within the body. The study follows a new Radio Frequency (RF) approach, by using a Vector Network Analyzer to provide a multifrequency measurement between 10 kHz and 1 GHz. A plastic and a metallic object were positioned within a saline-filled phantom to test the EIT reconstructed images at different frequencies. The results show that based on the image reconstruction, the material's conductivity can be distinguished at specific frequencies. However, beyond 10 MHz, the reconstructed images become less reliable until reaching 1 GHz, where the reconstructed image deteriorates significantly. While Electrical Impedance Tomography (EIT) has potential in new applications, its efficacy is limited by the ill-posed problem of the reconstruction. Further development and refinement of techniques, particularly in addressing issues that occur at higher frequencies, are required to realize EIT's full potential in various fields. [ABSTRACT FROM AUTHOR]

Trustworthy artificial intelligence (Trusted AI) is of utmost importance when learning-enabled components (LECs) are used in autonomous, safety-critical systems. When reliant on deep learning, these systems need to address the reliability, robustness, and interpretability of learning models. In addition to developing strategies to address these concerns, appropriate software architectures are needed to coordinate LECs and ensure they deliver acceptable behavior even under uncertain conditions. This work describes Anunnaki, a model-driven framework comprising loosely-coupled modular services designed to monitor and manage LECs with respect to Trusted AI assurance concerns when faced with different sources of uncertainty. More specifically, the Anunnaki framework supports the composition of independent, modular services to assess and improve the resilience and robustness of AI systems. The design of Annunaki was guided by several key software engineering principles (e.g., modularity, composability, and reusability) in order to facilitate its use and maintenance to support different aggregate monitoring and assurance analysis tools for LESs and their respective data sets. We demonstrate Anunnaki on two autonomous platforms, a terrestrial rover, and an unmanned aerial vehicle. Our studies show how Anunnaki can be used to manage the operations of different autonomous learning-enabled systems with vision-based LECs while exposed to uncertain environmental conditions. [ABSTRACT FROM AUTHOR]

MLOps is essential to streamline the machine learning (ML) development process, ensure ML models stay operational, and provide users with the desired value. MLOps enhances the auditability, dependability, repeatability, and quality of ML data, models, and systems. MLOps technologies tackle several operational difficulties in an ML process. This research used the TOE framework to identify drivers and challenges to adopting MLOps tool. Data were collected from 277 professionals from various industries and AI/ML-related job roles. The responses were analysed using a three-step approach – Data Profiling, Chi-square tests and Logistic regression (LR) model. The analysis uncovered that ML usage, performance drivers, and security drive MLOps adoption, whereas regulatory environment, organizational preparation, and ML infrastructure moderately influence it. The investigation shows that management/leadership needs to be aware of MLOps technologies' benefits. This study provides insights to AI/ML professionals, academics, researchers, and machine learning model users on MLOps adoption. [ABSTRACT FROM AUTHOR]

The article presents an invisible black-box backdoor attack against deep cross-modal hashing retrieval, addressing its vulnerability to backdoor attacks and the challenges posed by multi-modal data and hash quantization. Topics include developing a trigger generator to craft imperceptible triggers, designing an injection network for embedding triggers into benign samples, and proposing strategies to mitigate the impact of hash quantization on attack performance.

The integration of 6G networks with emerging key technologies such as blockchain, artificial intelligence, and digital twins continues to improve. However, it carries many issues with security threats and challenges of 6G networks. In this article, we analyzed the security issues of 6G networks and presented some possible solutions. First, we discussed the developments of mobile communication technology, the research motivation of 6G networks, the comparison of Key Performance Indicators (KPIs) between 5G/6G networks, and the key technologies of 6G networks. Next, security threats in the 6G network were analyzed concerning architecture, major visions and related applications. This was followed by solutions to security issues in applying key technologies for 6G networks. We also presented the application of AI in solving 6G network security problems. Firstly, we illustrated the impact of AI on 6G networks from two aspects: AI promotes the construction of 6G networks, and AI brings security threats to 6G networks. Then, we demonstrated that AI can assist 6G networks in solving security problems in many ways. Lastly, the is summarized, and the future directions in this area are proposed. [ABSTRACT FROM AUTHOR]

In this article, we present methods for the use of eye-tracking in interviews in order to reflect on visual perception and practices of looking as part of an ethnography of the senses. The methods are based on two multi-year ethnographic studies involving eye trackers. In the first one, researchers used mobile eye trackers to study how art museum visitors approach digital image technologies. In the other, they relied on stationary eye trackers to investigate practices on digital image platforms. We discuss how video recordings of participants' eye movements were made and describe the process of conducting ethnographic interviews based on the videos. The eye-tracking interviews can be used 1. to make participants aware of and think about practices of looking; 2. to verbalize in dialogue sensory and interpretative processes regarding museum objects and digital image technologies; and 3. to surface individuals' aesthetic preferences and incorporated knowledge. [ABSTRACT FROM AUTHOR]

Users in Asian countries (where adoption of broadband Internet technology is more widespread than anywhere else in the world) harbor notions of computer privacy that differ somewhat from those that are prevalent in Europe and the US. The spate of viruses and worms that has laid bare the security inadequacies of Microsoft's operating system and application software has exacted a particularly heavy toll on computer users in Asia. However, as well as posing challenges to privacy, technology also offers new opportunities for increasing privacy. Doug Tygar, computer science professor at the University of California, Berkeley, examines recent advances in security and privacy research, and suggests that the Asian computing environment represents a uniquely suitable test bed for new privacy technologies. He also highlights the increased popularity in Asia of Linux, the open-source operating system--due in part to its stability, but also to the fact that users need not fear hidden "backdoors" that might compromise security. [ABSTRACT FROM AUTHOR]

Digital security in all its forms, whether focused on computer security, information security, or cyber security, has practices of modelling at its core. Over recent decades, the goals of security in a digital context have widened to cover not only the protection of data and computing resource but also the protection and enablement of people using technology. Alongside this, there has been a concomitant diversification of the types of modelling that are used in digital security. In recent years, modellers have faced significant epistemological challenges, with tensions between different interdisciplinary perspectives about what models and modelling ought to be, and difficulties adequately communicating the virtues of different kinds of models among modellers and stakeholders such as policymakers or users. In this paper, we present a framework grounded in philosophy and social theory for understanding the methodological diversity of security modelling today: the MORS grid. Using the MORS grid, modellers can locate their own work of modelling, and explore methodological variations and political implications. We argue that in a subject that is increasingly recognized as being composed of different disciplinary positions, such a grid not only enables modellers to locate their own approach but also to appreciate the modelling positions of others. The grid is composed of four simple binaries: whether the modeller is an expert or non-expert, whether the modelling enquiry is model-oriented or target-oriented, whether the referent object (what needs to be protected) is determined prior to or posterior to the task of modelling, and whether the analyst adopts the design stance or the intentional stance. The paper presents the MORS grid through three lines of thought: first, we unpack the theoretical basis for each distinction in existing literature in the philosophy of science, security studies, and philosophy of mind; second, we provide a historical review of security modelling, and examine which positions on the MORS grid have predominated, and why; third, we set out the implications for modellers, policymakers and other stakeholders. [ABSTRACT FROM AUTHOR]

Privacy and security are complex topics, raising a variety of considerations and requirements that can be challenging to implement in software development. Determining the security and privacy related factors that have an influence on software systems development and deployment project outcomes has been the focus of extensive and ongoing research over the past two decades. To understand and categorize the factors that have an impact on developers' adoption and implementation of privacy and security considerations and practices in software development, we carried out a narrative review of the literature. The resulting mapping of factors provides a foundation for future interventions targeting organizational and individual behavior change, to increase the adoption of privacy and security practices in software development. [ABSTRACT FROM AUTHOR]

The ever-increasing number of computing devices around us results in more and more systems competing for our attention, making cognitive workload a crucial factor for the user experience of human-computer interfaces. Research in Human-Computer Interaction (HCI) has used various metrics to determine users' mental demands. However, there needs to be a systematic way to choose an appropriate and effective measure for cognitive workload in experimental setups, posing a challenge to their reproducibility. We present a literature survey of past and current metrics for cognitive workload used throughout HCI literature to address this challenge. By initially exploring what cognitive workload resembles in the HCI context, we derive a categorization supporting researchers and practitioners in selecting cognitive workload metrics for system design and evaluation. We conclude with three following research gaps: (1) defining and interpreting cognitive workload in HCI, (2) the hidden cost of the NASA-TLX, and (3) HCI research as a catalyst for workload-aware systems, highlighting that HCI research has to deepen and conceptualize the understanding of cognitive workload in the context of interactive computing systems. [ABSTRACT FROM AUTHOR]

Nonverbal sound offers great potential to enhance robots' interactions with humans, and a growing body of research has begun to explore nonverbal sound for tasks such as sound source localization, explicit communication, and improving sociability. However, nonverbal sound has a broad interpretation and design space that can draw from areas such as machine learning, music theory, and foley. We sought to identify and compare use cases and approaches for nonverbal sound in human-robot interaction through a systematic review. A search of sound and robotics-related publisher databases yielded 148 peer-reviewed articles presenting systems, studies, and taxonomies. Differences in taxonomy and overlap of terminology with adjacent research fields such as speech, gaze, and gesture posed difficulties for the search, which we attempted to address through a multi-stage search process. Based on the reviewed articles, we developed a pair of taxonomies using scientific communication principles and analyzed study designs and measures for the creation of nonverbal robot sound. We discuss recommendations for the field, including the use of the new taxonomies; methods for design, generation, and validation; and paths for future research. Roboticists may benefit from incorporating nonverbal sound as a key component in multimodal human-robot interaction. [ABSTRACT FROM AUTHOR]

The Controller Area Network (CAN) is the most widely used in-vehicle communication protocol, which still lacks the implementation of suitable security mechanisms such as message authentication and encryption. This makes the CAN bus vulnerable to numerous cyber attacks. Various Intrusion Detection Systems (IDSs) have been developed to detect these attacks. However, the high generalization capabilities of Artificial Intelligence (AI) make AI-based IDS an excellent countermeasure against automotive cyber attacks. This article surveys AI-based in-vehicle IDS from 2016 to 2022 (August) with a novel taxonomy. It reviews the detection techniques, attack types, features, and benchmark datasets. Furthermore, the article discusses the security of AI models, necessary steps to develop AI-based IDSs in the CAN bus, identifies the limitations of existing proposals, and gives recommendations for future research directions. [ABSTRACT FROM AUTHOR]

Although cyberattacks on machine learning (ML) production systems can be harmful, today, security practitioners are ill-equipped, lacking methodologies and tactical tools that would allow them to analyze the security risks of their ML-based systems. In this article, we perform a comprehensive threat analysis of ML production systems. In this analysis, we follow the ontology presented by NIST for evaluating enterprise network security risk and apply it to ML-based production systems. Specifically, we (1) enumerate the assets of a typical ML production system, (2) describe the threat model (i.e., potential adversaries, their capabilities, and their main goal), (3) identify the various threats to ML systems, and (4) review a large number of attacks, demonstrated in previous studies, which can realize these threats. To quantify the risk posed by adversarial machine learning (AML) threat, we introduce a novel scoring system that assigns a severity score to different AML attacks. The proposed scoring system utilizes the analytic hierarchy process (AHP) for ranking--with the assistance of security experts--various attributes of the attacks. Finally, we developed an extension to the MulVAL attack graph generation and analysis framework to incorporate cyberattacks on ML production systems. Using this extension, security practitioners can apply attack graph analysis methods in environments that include ML components thus providing security practitioners with a methodological and practical tool for both evaluating the impact and quantifying the risk of a cyberattack targeting ML production systems. [ABSTRACT FROM AUTHOR]

The prosperity of machine learning has been accompanied by increasing attacks on the training process. Among them, poisoning attacks have become an emerging threat during model training. Poisoning attacks have profound impacts on the target models, e.g., making them unable to converge or manipulating their prediction results. Moreover, the rapid development of recent distributed learning frameworks, especially federated learning, has further stimulated the development of poisoning attacks. Defending against poisoning attacks is challenging and urgent. However, the systematic review from a unified perspective remains blank. This survey provides an in-depth and up-to-date overview of poisoning attacks and corresponding countermeasures in both centralized and federated learning. We firstly categorize attack methods based on their goals. Secondly, we offer detailed analysis of the differences and connections among the attack techniques. Furthermore, we present countermeasures in different learning framework and highlight their advantages and disadvantages. Finally, we discuss the reasons for the feasibility of poisoning attacks and address the potential research directions from attacks and defenses perspectives, separately. [ABSTRACT FROM AUTHOR]

ICT platforms for news production, distribution, and consumption must exploit the ever-growing availability of digital data. These data originate from different sources and in different formats; they arrive at different velocities and in different volumes. Semantic knowledge graphs (KGs) is an established technique for integrating such heterogeneous information. It is therefore well-aligned with the needs of news producers and distributors, and it is likely to become increasingly important for the news industry. This article reviews the research on using semantic knowledge graphs for production, distribution, and consumption of news. The purpose is to present an overview of the field; to investigate what it means; and to suggest opportunities and needs for further research and development. [ABSTRACT FROM AUTHOR]

Machine learning (ML) has been universally adopted for automated decisions in a variety of fields, including recognition and classification applications, recommendation systems, natural language processing, and so on. However, in light of high expenses on training data and computing resources, recent years have witnessed a rapid increase in outsourced ML training, either partially or completely, which provides vulnerabilities for adversaries to exploit. A prime threat in training phase is called poisoning attack, where adversaries strive to subvert the behavior of machine learning systems by poisoning training data or other means of interference. Although a growing number of relevant studies have been proposed, the research among poisoning attack is still overly scattered, with each paper focusing on a particular task in a specific domain. In this survey, we summarize and categorize existing attack methods and corresponding defenses, as well as demonstrate compelling application scenarios, thus providing a unified framework to analyze poisoning attacks. Besides, we also discuss the main limitations of current works, along with the corresponding future directions to facilitate further researches. Our ultimate motivation is to provide a comprehensive and self-contained survey of this growing field of research and lay the foundation for a more standardized approach to reproducible studies. [ABSTRACT FROM AUTHOR]

In recent years, machine learning has transitioned from a field of academic research interest to a field capable of solving real-world business problems. However, the deployment of machine learning models in production systems can present a number of issues and concerns. This survey reviews published reports of deploying machine learning solutions in a variety of use cases, industries, and applications and extracts practical considerations corresponding to stages of the machine learning deployment workflow. By mapping found challenges to the steps of the machine learning deployment workflow, we show that practitioners face issues at each stage of the deployment process. The goal of this article is to lay out a research agenda to explore approaches addressing these challenges. [ABSTRACT FROM AUTHOR]

Adversarial attacks against supervised learninga algorithms, which necessitates the application of logging while using supervised learning algorithms in software projects. Logging enables practitioners to conduct postmortem analysis, which can be helpful to diagnose any conducted attacks. We conduct an empirical study to identify and characterize log-related coding patterns, i.e., recurring coding patterns that can be leveraged to conduct adversarial attacks and needs to be logged. A list of log-related coding patterns can guide practitioners on what to log while using supervised learning algorithms in software projects. We apply qualitative analysis on 3,004 Python files used to implement 103 supervised learning-based software projects. We identify a list of 54 log-related coding patterns that map to six attacks related to supervised learning algorithms. Using Log Assistant to conduct Postmortems for Supervised Learning (LOPSUL), we quantify the frequency of the identified log-related coding patterns with 278 open-source software projects that use supervised learning. We observe log-related coding patterns to appear for 22% of the analyzed files, where training data forensics is the most frequently occurring category. [ABSTRACT FROM AUTHOR]

Supporting users with secure password creation is a well-explored yet unresolved research topic. A promising intervention is the password meter, i.e. providing feedback on the user's password strength as and when it is created. However, findings related to the password meter's effectiveness are varied. An extensive literature review revealed that, besides password feedback, effective password meters often include: (a) feedback nudges to encourage stronger passwords choices and (b) additional guidance. A between-subjects study was carried out with 645 participants to test nine variations of password meters with different types of feedback nudges exploiting various heuristics and norms. This study explored differences in resulting passwords: (1) actual strength, (2) memorability, and (3) user perceptions. The study revealed that password feedback, in combination with a feedback nudge and additional guidance, labelled a hybrid password meter, was generally more efficacious than either intervention on its own, on all three metrics. Yet, the type of feedback nudge targeting either the person, the password creation task, or the social context, did not seem to matter much. The meters were nearly equally efficacious. Future work should explore the long-term effects of hybrid password meters in real-life settings to confirm the external validity of these findings. [ABSTRACT FROM AUTHOR]

Hardware security is currently a very influential domain, where each year countless works are published concerning attacks against hardware and countermeasures. A significant number of them use machine learning, which is proven to be very effective in other domains. This survey, as one of the early attempts, presents the usage of machine learning in hardware security in a full and organized manner. Our contributions include classification and introduction to the relevant fields of machine learning, a comprehensive and critical overview of machine learning usage in hardware security, and an investigation of the hardware attacks against machine learning (neural network) implementations. [ABSTRACT FROM AUTHOR]

Multi-tenant applications have been proliferating in recent years, supported by the emergence of computingas-service paradigms. Unfortunately, multi-tenancy induces new security vulnerabilities due to spatial or temporal co-location of applications with possibly malicious intent. In this article, we consider a special class of stealthy integrity attacks on multi-tenant deep learning accelerators. One interesting conclusion is that it is possible to perform targeted integrity attacks on kernel weights of deep learning systems such that it remains functional but mis-labels specific categories of input data through standard RowHammer attacks by only changing 0.0009% of the total weights. We develop an automated framework, AroMa, to evaluate the impact of multi-tenancy on security of deep learning accelerators against integrity attacks on memory systems. We present extensive evaluations on AroMa to demonstrate its effectiveness. [ABSTRACT FROM AUTHOR]

This article gives an in-depth presentation of the omni-big-step and omni-small-step styles of semantic judgments. These styles describe operational semantics by relating starting states to sets of outcomes rather than to individual outcomes. A single derivation of these semantics for a particular starting state and program describes all possible nondeterministic executions (hence the name omni), whereas in traditional small-step and big-step semantics, each derivation only talks about one single execution. This restructuring allows for straightforward modeling of both nondeterminism and undefined behavior as commonly encountered in sequential functional and imperative programs. Specifically, omnisemantics inherently assert safety (i.e., they guarantee that none of the execution branches gets stuck), while traditional semantics need either a separate judgment or additional error markers to specify safety in the presence of nondeterminism. Omnisemantics can be understood as an inductively defined weakest-precondition semantics (or more generally, predicate-transformer semantics) that does not involve invariants for loops and recursion but instead uses unrolling rules like in traditional small-step and big-step semantics. Omnisemantics were previously described in association with several projects, but we believe the technique has been underappreciated and deserves a well-motivated, extensive, and pedagogical presentation of its benefits. We also explore several novel aspects associated with these semantics, in particular, their use in type-safety proofs for lambda calculi, partial-correctness reasoning, and forward proofs of compiler correctness for terminating but potentially nondeterministic programs being compiled to nondeterministic target languages. All results in this article are formalized in Coq. [ABSTRACT FROM AUTHOR]

Existing cross domain recommender systems typically assume homogeneous user preferences across multiple domains to capture similarities of user-item interactions and to provide cross domain recommendations accordingly. Meanwhile, the heterogeneity of user behaviors is usually not well studied and captured during the recommendation process, where users might have vastly different interests in different domains. In addition, previous models focus primarily on recommendation tasks between domain pairs, and cannot be naturally extended to serve for multiple domain recommendation applications. To address these challenges, we propose to utilize the idea of adversarial learning to intelligently incorporate global user preferences and domain-specific user preferences for providing satisfying cross domain recommendations. In particular, our proposed Adversarial Cross Domain Recommendation (ACDR) model first obtains the latent representations of global user preferences from their explicit feature information, and then transforms them into domain-specific user embeddings, where we take into account user behaviors and their heterogeneous preferences among different domains. By doing so, we address the differences among user representations in the domain-specific latent space while also preserving global user preferences, as we effectively segment the distributions of domain-specific user embeddings in the shared latent space. The convergence of our proposed model is theoretically guaranteed. The proposed ACDR model leads to significant and consistent improvements in cross domain recommendation performance over the state-of-the-art baseline models, which we demonstrate through extensive experiments on three real-world datasets. In addition, we show that the improvements are greater on those datasets that are smaller and more sparse, on those users that have fewer interaction records in the dataset, and when user interactions from more product domains are included in the cross domain recommendation model. [ABSTRACT FROM AUTHOR]

In the past few decades, artificial intelligence (AI) technology has experienced swift developments, changing everyone's daily life and profoundly altering the course of human society. The intention behind developing AI was and is to benefit humans by reducing labor, increasing everyday conveniences, and promoting social good. However, recent research and AI applications indicate that AI can cause unintentional harm to humans by, for example, making unreliable decisions in safety-critical scenarios or undermining fairness by inadvertently discriminating against a group or groups. Consequently, trustworthy AI has recently garnered increased attention regarding the need to avoid the adverse effects that AI could bring to people, so people can fully trust and live in harmony with AI technologies. A tremendous amount of research on trustworthy AI has been conducted and witnessed in recent years. In this survey, we present a comprehensive appraisal of trustworthy AI from a computational perspective to help readers understand the latest technologies for achieving trustworthy AI. Trustworthy AI is a large and complex subject, involving various dimensions. In this work, we focus on six of the most crucial dimensions in achieving trustworthy AI: (i) Safety & Robustness, (ii) Nondiscrimination & Fairness, (iii) Explainability, (iv) Privacy, (v) Accountability & Auditability, and (vi) Environmental Well-being. For each dimension, we review the recent related technologies according to a taxonomy and summarize their applications in real-world systems. We also discuss the accordant and conflicting interactions among different dimensions and discuss potential aspects for trustworthy AI to investigate in the future. [ABSTRACT FROM AUTHOR]

Malicious software (malware) is a major cyber threat that has to be tackled with Machine Learning (ML) techniques because millions of new malware examples are injected into cyberspace on a daily basis. However, ML is vulnerable to attacks known as adversarial examples. In this article, we survey and systematize the field of Adversarial Malware Detection (AMD) through the lens of a unified conceptual framework of assumptions, attacks, defenses, and security properties. This not only leads us to map attacks and defenses to partial order structures, but also allows us to clearly describe the attack-defense arms race in the AMD context. We draw a number of insights, including: knowing the defender’s feature set is critical to the success of transfer attacks; the effectiveness of practical evasion attacks largely depends on the attacker’s freedom in conducting manipulations in the problem space; knowing the attacker’s manipulation set is critical to the defender’s success; and the effectiveness of adversarial training depends on the defender’s capability in identifying the most powerful attack. We also discuss a number of future research directions. [ABSTRACT FROM AUTHOR]

Geographical information systems (GIS) are essential tools for mineral prospectivity modeling (MPM). Three-dimensional (3D) MPM is able to learn the association between geological evidence and mineralization in shallow zones and thereby build a prospectivity model for deep zones, making it a desirable technique to target deep-seated orebodies. However, existing 3D MPM methods directly generalize the model learned in shallow zones to the deep zones without attention to model transferability caused by the different metallogenic mechanisms between the two zones. In this study, we aim to robustly transfer the prospectivity model learned from shallow zones to deep zones. We cast the 3D MPM as a domain adaptation problem, which is an important realm of transfer learning. Because the metallogenic mechanism can be closely associated with spatial locations, we specifically focus on domain adaption concerning the spatial locations that are ignored by conventional domain adaptation methods. To measure the spatial-associated domain discrepancy, we propose a novel spatial-associated maximum mean discrepancy (SAMMD), which compares the joint distributions of features and spatial locations across domains. Based on the SAMMD criterion, a deep neural network, referred to as the spatial-associated domain adaptation network, is devised to learn cross-domain but mineralization-indicative features for building prospectivity model that is transferable to deep zones. A case study of the world-class Sanshandao gold deposit, in eastern China, was carried out to validate the effectiveness of the proposed methods. The results show that compared with other leading MPM methods and other domain adaption variants, the proposed method has superior prediction accuracy and targeting efficiency, demonstrating the effectiveness and robustness of the proposed method in targeting deep-seated orebodies in areas with different metallogenic mechanisms and no labeled data. [ABSTRACT FROM AUTHOR]

The incremental diffusion of machine learning algorithms in supporting cybersecurity is creating novel defensive opportunities but also new types of risks. Multiple researches have shown that machine learning methods are vulnerable to adversarial attacks that create tiny perturbations aimed at decreasing the effectiveness of detecting threats. We observe that existing literature assumes threat models that are inappropriate for realistic cybersecurity scenarios, because they consider opponents with complete knowledge about the cyber detector or that can freely interact with the target systems. By focusing on Network Intrusion Detection Systems based on machine learning, we identify and model the real capabilities and circumstances required by attackers to carry out feasible and successful adversarial attacks. We then apply our model to several adversarial attacks proposed in literature and highlight the limits and merits that can result in actual adversarial attacks. The contributions of this article can help hardening defensive systems by letting cyber defenders address the most critical and real issues and can benefit researchers by allowing them to devise novel forms of adversarial attacks based on realistic threat models. [ABSTRACT FROM AUTHOR]

In the last decades, research has shown that both technical solutions and user perceptions are important to improve security and privacy in the digital realm. The field of ‘usable security’ already started to emerge in the mid-90s, primarily focussed on password and email security. Later on, the research field of ”usable security and privacy” evolved and broadened the aim to design concepts and tools to assist users in enhancing their behaviour with regard to both privacy and security. Nevertheless, many user interventions are not as effective as desired. Because of highly diverse usage contexts, leading to different privacy and security requirements and not always to one-size-fits-all approaches, tailorability is necessary to address this issue. Furthermore, transparency is a crucial requirement, as providing comprehensible information may counter reactance towards security interventions. This article first provides a brief history of the research field in its first quarter-century and then highlights research on the transparency and tailorability of user interventions. Based on this, this article then presents six contributions with regard to (1) privacy concerns in times of COVID-19, (2) authentication on mobile devices, (3) GDPR-compliant data management, (4) privacy notices on websites, (5) data disclosure scenarios in agriculture, as well as (6) rights under data protection law and the concrete process should data subjects want to claim those rights. This article concludes with several research directions on user-centred transparency and tailorability. [ABSTRACT FROM AUTHOR]

In recent years, machine learning algorithms, and more specifically deep learning algorithms, have been widely used in many fields, including cyber security. However, machine learning systems are vulnerable to adversarial attacks, and this limits the application of machine learning, especially in non-stationary, adversarial environments, such as the cyber security domain, where actual adversaries (e.g., malware developers) exist. This article comprehensively summarizes the latest research on adversarial attacks against security solutions based on machine learning techniques and illuminates the risks they pose. First, the adversarial attack methods are characterized based on their stage of occurrence, and the attacker' s goals and capabilities. Then, we categorize the applications of adversarial attack and defense methods in the cyber security domain. Finally, we highlight some characteristics identified in recent research and discuss the impact of recent advancements in other adversarial learning domains on future research directions in the cyber security domain. To the best of our knowledge, this work is the first to discuss the unique challenges of implementing end-to-end adversarial attacks in the cyber security domain, map them in a unified taxonomy, and use the taxonomy to highlight future research directions. [ABSTRACT FROM AUTHOR]

Malicious software (malware) is a major cyber threat that has to be tackled with Machine Learning (ML) techniques because millions of new malware examples are injected into cyberspace on a daily basis. However, ML is vulnerable to attacks known as adversarial examples. In this article, we survey and systematize the field of Adversarial Malware Detection (AMD) through the lens of a unified conceptual framework of assumptions, attacks, defenses, and security properties. This not only leads us to map attacks and defenses to partial order structures, but also allows us to clearly describe the attack-defense arms race in the AMD context. We draw a number of insights, including: knowing the defender’s feature set is critical to the success of transfer attacks; the effectiveness of practical evasion attacks largely depends on the attacker’s freedom in conducting manipulations in the problem space; knowing the attacker’s manipulation set is critical to the defender’s success; and the effectiveness of adversarial training depends on the defender’s capability in identifying the most powerful attack. We also discuss a number of future research directions. [ABSTRACT FROM AUTHOR]

We conduct a comprehensive review covering academic publications and industry products relating to tools for cybersecurity awareness and education aimed at non-expert end-users developed in the past 20 years. Through our search criteria, we identified 119 tools that we cataloged into five broad media categories. We explore current trends, assess their use of relevant instructional design principles, and review empirical evi)dence of the tools’ effectiveness. From our review, we provide an evaluation checklist and suggest that a more systematic approach to the design and evaluation of cybersecurity educational tools would be beneficial. [ABSTRACT FROM AUTHOR]

In this paper, we propose a new approach for classifying malignant moles. From each image, we extract ten different types of features. All attributes are inserted into multi intelligent techniques that give a very high rating among them. [ABSTRACT FROM AUTHOR]