A Bill of Materials (BoM) is a list of all components on a printed circuit board (PCB). Since BoMs are useful for hardware assurance, automatic BoM extraction (AutoBoM) is of great interest to the government and electronics industry. To achieve a high-accuracy AutoBoM process, domain knowledge of PCB text and logos must be utilized. In this study, we discuss the challenges associated with automatic PCB marking extraction and propose 1) a plan for collecting salient PCB marking data, and 2) a framework for incorporating this data for automatic PCB assurance. Given the proposed dataset plan and framework, subsequent future work, implications, and open research possibilities are detailed.

Artificial intelligence (AI) and machine learning (ML) techniques have been increasingly used in several fields to improve performance and the level of automation. In recent years, this use has exponentially increased due to the advancement of high-performance computing and the ever increasing size of data. One of such fields is that of hardware design—specifically the design of digital and analog integrated circuits, where AI/ ML techniques have been extensively used to address ever-increasing design complexity, aggressive time to market, and the growing number of ubiquitous interconnected devices. However, the security concerns and issues related to integrated circuit design have been highly overlooked. In this article, we summarize the state-of-the-art in AI/ML for circuit design/optimization, security and engineering challenges, research in security-aware computer-aided design/electronic design automation, and future research directions and needs for using AI/ML for security-aware circuit design.

Abstract---Reinforcement learning (RL) has become more popular due to promising results in applications such as chat-bots, healthcare, and autonomous driving. However, one significant challenge in current RL research is the difficulty in understanding which RL algorithms, if any, are practical for a given use case. Few RL algorithms are rigorously tested, and hence understood, for their practical implications. Although there are a number of performance comparisons in literature, many use few environments and do not consider real-world limitations such as run-time and memory usage. Furthermore, many works do not make their code publicly accessible for others to use. This paper addresses this gap by presenting the most comprehensive performance comparison on the practicality of RL algorithms known to date. Specifically, this paper focuses on discrete, model-free deep RL algorithms for their practicality in real-world problems where efficient implementations are necessary. In total, fourteen RL algorithms were trained on twenty-three environments (468 environment instances), which collectively required 224 GB and 766 days CPU time to run all experiments, and 1.7 GB to store all models. Overall, the results indicate several shortcomings in RL algorithms' exploration efficiency, memory/sample efficiency, and space/time complexity. Based on these shortcomings, numerous opportunities for future works were identified to improve the capabilities of modern algorithms. This paper’s findings will help researchers and practitioners improve and employ RL algorithms in time-sensitive and resource-constrained applications such as economics, cybersecurity, and Internet of Things (IoT). Impact Statement---Reinforcement learning (RL) technologies are commonly used in autonomous driving, chat-bot, and business analytic applications. They learn how to adapt to unforeseen situations, reducing the load on human drivers, support teams, and analysts. Although there are a variety of theoretical works in RL literature, very few algorithms are tested and evaluated to facilitate their use in real-life scenarios. The performance comparison introduced in this paper addresses these limitations. The performance analysis framework, re-implemented source code, and findings identified in this study could increase the adoption and speed development of RL technologies in more real-life applications. Moreover, the open challenges, recommendations, and practical implications identified in this paper could facilitate collaboration and development of new technologies among researchers and practitioners in industry and academia.

Phishing and disinformation are popular social engineering attacks with attackers invariably applying influence cues in texts to make them more appealing to users. We introduce Lumen, a learning-based framework that exposes influence cues in text: (i) persuasion, (ii) framing, (iii) emotion, (iv) objectivity/subjectivity, (v) guilt/blame, and (vi) use of emphasis. Lumen was trained with a newly developed dataset of 3K texts comprised of disinformation, phishing, hyperpartisan news, and mainstream news. Evaluation of Lumen in comparison to other learning models showed that Lumen and LSTM presented the best F1-micro score, but Lumen yielded better interpretability. Our results highlight the promise of ML to expose influence cues in text, toward the goal of application in automatic labeling tools to improve the accuracy of human-based detection and reduce the likelihood of users falling for deceptive online content.

Most semantic image annotation platforms suffer severe bottlenecks when handling large images, complex regions of interest, or numerous distinct foreground regions in a single image. We have developed the Semi-Supervised Semantic Annotator (S3A) to address each of these issues and facilitate rapid collection of ground truth pixel-level labeled data. Such a feat is accomplished through a robust and easy-to-extend integration of arbitrary python image processing functions into the semantic labeling process. Importantly, the framework devised for this application allows easy visualization and machine learning prediction of arbitrary formats and amounts of per-component metadata. To our knowledge, the ease and flexibility offered are unique to S3A among all open-source alternatives.

We posit that the next major thrust relevant to capturing dynamics for detecting and responding to information operations is inference of hidden mental states through natural language processing and social computing techniques. An important factor contributing to this vision is the need for explainable representations, e.g., propositions, to capture hidden mental states as indicators of influence campaigns. Hidden mental states under exploration include, for example, belief, stance, and concern. We view explainability not in terms of a “reason describer” for machine learning (ML) model output, but in terms of an inherently interpretable paradigm that leverages hidden mental states to produce both an explanation and a justification of output. The aim is to reap the benefits of both worlds: (1) breadth of coverage for features that are essential to the task at hand (e.g., embedding and attention models for extracting sentiment); (2) depth and transparency of representational formalisms for explaining system decisions (e.g., propositions that identify beliefs and attitudes).

Reverse engineering (RE) is the only foolproof method of establishing trust and assurance in hardware. This is especially important in today's climate, where new threats are arising daily. A Printed Circuit Board (PCB) serves at the heart of virtually all electronic systems and, for that reason, a precious target amongst attackers. Therefore, it is increasingly necessary to validate and verify these hardware boards both accurately and efficiently. When discussing PCBs, the current state-of-the-art is non-destructive RE through X-ray Computed Tomography (CT); however, it remains a predominantly manual process. Our work in this paper aims at paving the way for future developments in the automation of PCB RE by presenting automatic detection of vias, a key component to every PCB design. We provide a via detection framework that utilizes the Hough circle transform for the initial detection, and is followed by an iterative false removal process developed specifically for detecting vias. We discuss the challenges of detecting vias, our proposed solution, and lastly, evaluate our methodology not only from an accuracy perspective but the insights gained through iteratively removing false-positive circles as well. We also compare our proposed methodology to an off-the-shelf implementation with minimal adjustments of Mask R-CNN; a fast object detection algorithm that, although is not optimized for our application, is a reasonable deep learning model to measure our work against. The Mask R-CNN we utilize is a network pretrained on MS COCO followed by fine tuning/training on prepared PCB via images. Finally, we evaluate our results on two datasets, one PCB designed in house and another commercial PCB, and achieve peak results of 0.886, 0.936, 0.973, for intersection over union (IoU), Dice Coefficient, and Structural Similarity Index. These results vastly outperform our tuned implementation of Mask R-CNN.

Phishing is fundamental to cyber attacks. This research determined the effect of Internet user age and email content such as weapons of influence (persuasive techniques that attackers can use to lure individuals to fall for an attack) and life domains (a specific topic or aspect of an individual's life that attackers can focus an email on) on spear-phishing (targeted phishing) susceptibility. In total, 100 young and 58 older users received, without their knowledge, daily simulated phishing emails over 21 days. A browser plugin recorded their clicking on links in the emails as an indicator of their susceptibility. Forty-three percent of users fell for the simulated phishing emails, with older women showing the highest susceptibility. While susceptibility in young users declined across the study, susceptibility in older users remained stable. The relative effectiveness of the attacks differed by weapons of influence and life domains with age-group variability. In addition, older compared to young users reported lower susceptibility awareness. These findings support effects of Internet user demographics and email content on susceptibility to phishing and emphasize the need for personalization of the next generation of security solutions.

System-on-chips (SoCs) are increasingly used in high assurance electronic systems such as military, space, automotive, financial and health care systems. We rely on the security of these SoCs, yet it has been demonstrated that SoCs can be compromised using physical and remote attacks such as timing and power analysis or fault-injection exploits. Manually analyzing circuit implementations for security vulnerabilities is becoming less and less feasible due to the increasing complexity of SoC designs and the integration of multiple third-party IP (3PIP) blocks. Security policies also lack formal definitions and rely on ad-hoc solutions at design time. Currently, there exists no systematic, automated solution to deal with SoC vulnerabilities before deployment. To address this, we propose the unified security definition format (USDF) language for the formal representation of IC security policies and attributes. We extend the design security rule check (DSeRC) framework and introduce the automated security policy enforcement (ASPEN) framework. USDF provides a language syntax to formally write security policies and attributes that can be interpreted by the ASPEN framework for automated SoC security analysis. Keywords—system on chip; security policy

We demonstrate bidirectional continuous electrowetting by embedding metal–semiconductor diodes in the electrowetting substrate. Unlike conventional electrowetting on dielectric, bidirectional continuous electrowetting uses a single electrode pair to actuate a droplet through long distances. As long as the voltage potential is maintained between two end electrodes, the droplet moves toward the electrode with the higher potential. However, previously reported material systems had limited success in repeated actuation. In this work, diodes based on Schottky barriers were fabricated by forming metal–semiconductor junctions between titanium and high-resistivity n-type silicon. The performance enhancements were evaluated using current–voltage measurements of interface pairs. When the titanium is coated with gold to limit electrochemical reactions, the Schottky diodes achieved superior performance compared to electrochemical diodes previously studied. Droplet speed range from 8 to 240 mm/s is reported. Under repeated actuation, the speed of the droplet showed no degradation for up to 2000 cycles (experiment duration).

Microfluidic-driven mechanical actuation opens new possibilities for positioning and manipulating delicate small components. However, existing microfluidic actuation methods are not well-suited to positioning with high resolution. This paper reports a method for precise, open-loop control of droplet position in finite steps by varying the duty cycle of the input signal in electrowetting actuation. When wetted to a solid object, both the droplet and the solid can be actuated. Unlike conventional electrowetting actuation methods, positioning resolution in our proposed method can be much smaller than the size of the underlying electrodes without requiring closed loop feedback control system. Using a leaky dielectric coating, the electrode/electrolyte combination in our device acts as a simple diode by blocking current in one direction and conducting in the other. Each duty cycle of the applied AC square wave corresponds to a unique position on the electrode. The position - duty cycle relationship is found to be nonlinear but symmetric about the center of the electrodes. This approach provides a method for improving open-loop positioning resolution without adding more electrodes. Positioning is within 0.2 mm (<2.5% of the droplet diameter) of the idealized model and repeatability is <0.07 mm (<0.8% of the droplet diameter). (C) 2016 Elsevier B.V. All rights reserved.

This paper illustrates the dynamic effects of using a kinetic shape as a crutch tip on swing through crutch walking (non-weight bearing). The overground crutch walking of four participants was measured to examine the effect of a Kinetic Crutch Tip (KCT) on step length and swing time using a ProtoKinetics ® Zeno Walkway System. Changes in ground reaction forces during the crutch gait cycle were examined by having the participants walk on an instrumented treadmill. We quantify changes in crutch dynamics by comparing results to standard rubber point tip crutch walking. The results showed that introducing a KCT to crutch walking can alter step length and swing time asymmetries during overground walking. Participants walking with a forward forcing KCT experienced a reduction in the horizontal ground reaction forces of up to 74% compared to walking on standard rubber crutch tips. The backward forcing KCT reduced the heel strike peak forces by as much as 27%. These findings show that crutch walking dynamics can be customized and optimized to yield a specific crutch walking behavior tailored to various user needs or walking environments.

Electrowetting on dielectric is a phenomenon in which the shape and apparent contact angle of a droplet changes when an electric field is applied across the droplet interface. If the field is asymmetric with respect to the droplet, then a net force can be applied to the droplet. In this work, we have measured the electrowetting force by confining the droplet shape beneath a glass plate and measuring the force on the plate. The force was measured as a function of voltage for a range of fluids with different surface energy. Measured forces show excellent agreement with predictions based on the Young–Lippmann equation with measured contact angles. Results also show that the electrowetting force is independent of fluid surface energy below saturation but that the peak force is proportional to the surface tension. This work shows that lowering the surface energy of the fluid can induce larger contact angle change under the same voltage, but it has no beneficial impact on the actuation force in droplet-based actuators. In contrast, velocity tests with deformable droplets show higher speeds for lower surface energy fluids, even above their saturation voltage. However, when the droplet’s shape is restrained, the highest velocity is achieved with high surface energy fluids due to the larger electrowetting actuation forces applied.