Promise of AI in Defi, A Literary Analysis

Decentralized Finance (DeFi) is an emerging and revolutionizing field with notable uncertainties of reliability to be used on a mass scale. On the other hand, Artificial Intelligence (AI) has proved to be a crucial helping tool in numerous domains. In this study, we present a systematic review of the utility of AI in Defi in terms of impact, reliability, and security and conduct exhaustive analysis. We further conclude from our extensive literature review that we can identify possible new research opportunities in AI to bridge the gap of trust between peers and make the integration of DeFi more agile in the near future.

Property Evolution and Reliability of Underfills Under Sustained High Temperature Storage

Automotive underhood electronics may be exposed to high temperature in the neighborhood of 100°C–200°C. Property evolution may impact reliability and accuracy of predictive models to assure desired use life. In this paper, evolution of properties of two underfill material properties are studied using DMA (Dynamic Mechanical Analyzer). The underfills are exposed to three different operational temperatures in the range of 100°C to 140°C for the measurements. The dynamic mechanical properties such as storage modulus (E′), loss modulus (E″), tangent delta (tan δ), and respective glass transition temperatures (Tg) are studied using DMA. Study of viscoelastic behavior of underfills is achieved by performing TTS (time-temperature superposition) experiments at 7 discrete frequencies 0.1, 0.21, 0.46, 1, 2.15, 4.64, and 10 Hz using DMA in three-point bend mode. From the selected reference temperatures, the master curves were constructed for storage moduli, loss moduli and tan delta as a function of frequency using TTS results. Using the WLF (Williams-Landel-Ferry) equation, the shift factors as a function of temperature were determined along the frequency axis. The relaxation modulus as a function of temperature and time can be obtained using the master curves of storage and loss moduli. A simple and detailed procedure has been established to find the Prony series constants.

Promise of AI in Defi, A Literary Analysis

Decentralized Finance (DeFi) is an emerging and revolutionizing field with notable uncertainties of reliability to be used on a mass scale. On the other hand, Artificial Intelligence (AI) has proved to be a crucial helping tool in numerous domains. In this study, we present a systematic review of the utility of AI in Defi in terms of impact, reliability, and security and conduct exhaustive analysis. We further conclude from our extensive literature review that we can identify possible new research opportunities in AI to bridge the gap of trust between peers and make the integration of DeFi more agile in the near future.

Neural predictive monitoring and a comparison of frequentist and Bayesian approaches

Neural state classification (NSC) is a recently proposed method for runtime predictive monitoring of hybrid automata (HA) using deep neural networks (DNNs). NSC trains a DNN as an approximate reachability predictor that labels an HA state x as positive if an unsafe state is reachable from x within a given time bound, and labels x as negative otherwise. NSC predictors have very high accuracy, yet are prone to prediction errors that can negatively impact reliability. To overcome this limitation, we present neural predictive monitoring (NPM), a technique that complements NSC predictions with estimates of the predictive uncertainty. These measures yield principled criteria for the rejection of predictions likely to be incorrect, without knowing the true reachability values. We also present an active learning method that significantly reduces the NSC predictor’s error rate and the percentage of rejected predictions. We develop two versions of NPM based, respectively, on the use of frequentist and Bayesian techniques to learn the predictor and the rejection rule. Both versions are highly efficient, with computation times on the order of milliseconds, and effective, managing in our experimental evaluation to successfully reject almost all incorrect predictions. In our experiments on a benchmark suite of six hybrid systems, we found that the frequentist approach consistently outperforms the Bayesian one. We also observed that the Bayesian approach is less practical, requiring a careful and problem-specific choice of hyperparameters.

User-Adapted Web Services by Extending the eIDAS Specification with Functional Attributes

To provide web services adapted to the users’ functional capabilities, diversity must be considered from the conceptualization and design phases of the services’ development. In previous work, we proposed a model that allows the provisioning of adapted interfaces based on users’ identity and their functional attributes to facilitate this task to software designers and developers. However, being these identities and attributes self-declared by the users may impact reliability and usability. In this work, we propose an extension of our model to resolve these deficiencies by delegating the identity and attributes provision to external certified entities. The European electronic Identification, Authentication and Trust Services (eIDAS) regulation established a solution to ensure the cross-border mutual recognition of Electronic Identification (eID) mechanisms among the European Member States. This research aims to provide an extension of this regulation mentioned above (eIDAS) to support functional attributes and connect our previously proposed model to this extended eIDAS network. Thanks to this proposal, web services can guarantee adapted and personalized interfaces while improving the functionalities offered without any previous configuration by users and, in a reliable way, since the functional attributes belong to the users’ official eID. As the attributes set provided by eIDAS nodes only contains citizens’ personal and legal ones, we also propose a mechanism to connect the eIDAS network to external attribute providers that could extend the eIDAS profile of users with their functional attributes. We have deployed a pilot to validate the proposed model consisting of an identity provider, an eIDAS node supporting the extended reference code and an attribute provider supporting functional attributes. We have also designed and implemented a simple service that supports eID authentication and serves adapted interfaces based on the retrieved extended eIDAS profile. Finally, we have developed an experience for getting feedback from a set of real users with different functional capabilities. According to the results, we conclude that the generalized adoption of the proposed solution in the European digital web services will significantly improve their accessibility in terms of ease of use and adaptability to users’ capacities.

Non-Perpendicular Orientation High-G Impact Reliability of Electronics Potted Assemblies

The Commercial electronics being used in defense and aerospace applications are being exposed to extreme environments including high-G shock conditions, which is not their intended purpose of use. Currently most of the board level testing is being done at horizontal zero degree drop angle. In real life drop scenarios, the angle of drop varies a lot. The damage accrued in the board interconnects and components and solder-joint interconnects, varies with the change in the drop angle. The reliability of the electronic components and interconnections of the solder-joint depends on the effect of drop angle on the test vehicle. The results acquired under these varying drop angle environments would be more relatable to the real life drop scenarios. The test vehicle is a circular PCB and two different configurations of the test vehicle are tested bare and potted. The boards are tested for three different drop angles of 0-degree, 30-degree and 60 degree. Two different shock levels are tested at each drop angle 10,000g and 25,000g. To predict the effect of drop angle on the test assembly, an explicit finite element model of the assembly has been created and simulated.

Liquid-Cooled Heat Sink Optimization for Thermal Imbalance Mitigation in Wide-Bandgap Power Modules

Power semiconductor die layout in substrates used in power modules is generally optimized for minimum electrical parasitics (e.g., stray inductance) by considering the minimum spacing between dies for thermal decoupling. The layout assumes sufficient heat spreading and transfer from dies to the cooling structure. For module designs using a direct substrate cooling method, the base plate is removed, leading to a steady-state thermal asymmetry in the power module due to insufficient heat spreading/transfer. This causes significant temperature differences among the devices. Such unintentional thermal asymmetries can lead to undesirable asymmetries in power conversion among semiconductor devices, which impact reliability. This article proposes a thermal imbalance mitigation method that uses evolutionary optimized liquid-cooled heat sinks to improve the thermal loading among devices.