Contributions in Merlin

This paper estimates the curvature of the Earth, defined as one over its radius, without relying on physical measurements. The orthodox model states that the Earth is (nearly) spherical with a curvature of π/20,000km. By contrast, the heterodox flat-Earth model stipulates a curvature of zero. Abstracting from the well-worn arguments for and against both models, rebuttals and counter-rebuttals ad infinitum, we propose a novel statistical methodology based on verifiable flight times along regularly scheduled commercial airline routes; this methodology allows for both estimating and making inference for Earth’s curvature. In particular, a formal hypothesis test resolutely rejects the flat-Earth model, whereas it does not reject the orthodox spherical-Earth model.

In the last years, the number of IoT devices deployed has suffered an undoubted explosion, reaching the scale of billions. However, some new cybersecurity issues have appeared together with this development. Some of these issues are the deployment of unauthorized devices, malicious code modification, malware deployment, or vulnerability exploitation. This fact has motivated the requirement for new device identification mechanisms based on behavior monitoring. Besides, these solutions have recently leveraged Machine and Deep Learning (ML/DL) techniques due to the advances in this field and the increase in processing capabilities. In contrast, attackers do not stay stalled and have developed adversarial attacks focused on context modification and ML/DL evaluation evasion applied to IoT device identification solutions. However, literature has not yet analyzed in detail the impact of these attacks on individual identification solutions and their countermeasures. This work explores the performance of hardware behavior-based individual device identification, how it is affected by possible context- and ML/DL-focused attacks, and how its resilience can be improved using defense techniques. In this sense, it proposes an LSTM-CNN architecture based on hardware performance behavior for individual device identification. Then, the most usual ML/DL classification techniques have been compared with the proposed architecture using a hardware performance dataset collected from 45 Raspberry Pi devices running identical software. The LSTM-CNN improves previous solutions achieving a ＋0.96 average F1-Score and 0.8 minimum TPR for all devices. Afterward, context- and ML/DL-focused adversarial attacks were applied against the previous model to test its robustness. A temperature-based context attack was not able to disrupt the identification, but some ML/DL state-of-the-art evasion attacks were successful. Finally, adversarial training and model distillation defense techniques are selected to improve the model resilience to evasion attacks, improving its robustness from up to 0.88 attack success ratio to 0.17 in the worst attack case, without degrading its performance in an impactful manner.

Habits pose a fundamental puzzle for those aiming to understand human behavior. They pervade our everyday lives and dominate some forms of psychopathology but are extremely hard to elicit in the lab. In this Registered Report, we developed novel experimental paradigms grounded in computational models, which suggest that habit strength should be proportional to the frequency of behavior and, in contrast to previous research, independent of value. Specifically, we manipulated how often participants performed responses in two tasks varying action repetition without, or separately from, variations in value. Moreover, we asked how this frequency-based habitization related to value-based operationalizations of habit and self-reported propensities for habitual behavior in real life. We find that choice frequency during training increases habit strength at test and that this form of habit shows little relation to value-based operationalizations of habit. Our findings empirically ground a novel perspective on the constituents of habits and suggest that habits may arise in the absence of external reinforcement. We further find no evidence for an overlap between different experimental approaches to measuring habits and no associations with self-reported real-life habits. Thus, our findings call for a rigorous reassessment of our understanding and measurement of human habitual behavior in the lab.

We study the long-term effects of a randomized intervention targeting children's socio-emotional skills. The classroom-based intervention for primary school children has positive impacts that persist for over a decade. Treated children become more likely to complete academic high school and enrol in university. Two mechanisms drive these results. Treated children show fewer attention deficit/hyperactivity disorder symptoms: they are less impulsive and less disruptive. They also attain higher grades, but they do not score higher on standardized tests. The long-term effects on educational attainment thus appear to be driven by changes in socio-emotional skills rather than cognitive skills.

The computing device deployment explosion experienced in recent years, motivated by the advances of technologies such as Internet-of-Things (IoT) and 5G, has led to a global scenario with increasing cybersecurity risks and threats. Among them, device spoofing and impersonation cyberattacks stand out due to their impact and, usually, low complexity required to be launched. To solve this issue, several solutions have emerged to identify device models and types based on the combination of behavioral fingerprinting and Machine/Deep Learning (ML/DL) techniques. However, these solutions are not appropriate for scenarios where data privacy and protection are a must, as they require data centralization for processing. In this context, newer approaches such as Federated Learning (FL) have not been fully explored yet, especially when malicious clients are present in the scenario setup. The present work analyzes and compares the device model identification performance of a centralized DL model with an FL one while using execution time-based events. For experimental purposes, a dataset containing execution-time features of 55 Raspberry Pis belonging to four different models has been collected and published. Using this dataset, the proposed solution achieved 0.9999 accuracy in both setups, centralized and federated, showing no performance decrease while preserving data privacy. Later, the impact of a label-flipping attack during the federated model training is evaluated using several aggregation mechanisms as countermeasures. Zeno and coordinate-wise median aggregation show the best performance, although their performance greatly degrades when the percentage of fully malicious clients (all training samples poisoned) grows over 50%.

We need more computing capacity for the next growth cycle, and computers with conventional transistor technology are reaching their limits. So new ideas are required. The quantum computer, which overcomes the binary system and is not based on silicon microchips, could be a solution. This technology will continue to develop exponentially and transform science, the economy, and society. Furthermore, the paradigm of quantum communication offers an entirely novel possibility of distributed computing by allowing quantum computers to be networked via quantum channels to intrinsically secure communication. This article explains how quantum computers exploit new phenomena that do not occur in classical physics. Along the four primary application areas identified (optimization, simulation, machine learning, and cryptography), we describe possible applications in various industries. Our critical appraisal presents the technical challenges that still hold the potential for quantum computing to complement traditional computing systems. Accordingly, small and mid-sized companies do not necessarily need to invest in quantum computers but in their use. Quantum as a service can be the first step for visionary leaders to get familiar with it and gain a competitive advantage early on.

We assess the impact of the leverage ratio capital requirements on the risk-taking behaviour of banks both theoretically and empirically. Conceptually, introducing binding leverage ratio requirements into a regulatory framework with risk-based capital requirements induces banks to re-optimise, shifting from safer to riskier assets (higher asset risk). Yet, this shift would not be one-for-one due to risk weight differences, meaning the shift would be associated with a lower level of leverage (lower insolvency risk). The interaction of these two changes determines the impact on the aggregate level of risk. Empirically, we use a difference-in-differences setup to compare the behaviour of UK banks subject to the leverage ratio requirements (LR banks) to otherwise similar banks (non-LR banks). Our results show that LR banks did not increase asset risk, and slightly reduced leverage levels, compared to the control group after the introduction of leverage ratio in the UK. As expected, these two changes led to a lower aggregate level of risk. Emperical results indicate that credit default swap spreads on the 5-year subordinated debt of LR banks decreased relative to non-LR banks post leverage ratio introduction, suggesting the market viewed LR banks as less risky, especially during the COVID 19 stress.

This article presents a portfolio construction approach that combines the hierarchical clustering of a large asset universe with the stock price momentum. On one hand, investing in high-momentum stocks enhances returns by capturing the momentum premium. On the other hand, hierarchical clustering of a high-dimensional asset universe ensures sparse diversification, stabilizes the portfolio across economic regimes, and mitigates the problem of increased drawdowns typically present in momentum portfolios. Moreover, the proposed portfolio construction approach avoids the covariance matrix inversion. An out-of-sample backtest on a non-survivorship-biased dataset of international stocks shows that, compared to the model-based and model-free benchmarks, hierarchical momentum portfolios achieve improved cumulative and risk-adjusted portfolio returns as well as decreased portfolio drawdowns net of transaction costs. The study further suggests that the unique characteristics of the hierarchical momentum portfolios arise because of both dimensionality reduction via clustering and momentum-based stock selection.

The proliferation of the Internet of Things (IoT) has led to the emergence of crowdsensing applications, where a multitude of interconnected devices collaboratively collect and analyze data. Ensuring the authenticity and integrity of the data collected by these devices is crucial for reliable decision-making and maintaining trust in the system. Traditional authentication methods are often vulnerable to attacks or can be easily duplicated, posing challenges to securing crowdsensing applications. Besides, current solutions leveraging device behavior are mostly focused on device identification, which is a simpler task than authentication. To address these issues, an individual IoT device authentication framework based on hardware behavior fingerprinting and Transformer autoencoders is proposed in this work. To support the design, a threat model details the security problems faced when performing hardware-based authentication in IoT. This solution leverages the inherent imperfections and variations in IoT device hardware to differentiate between devices with identical specifications. By monitoring and analyzing the behavior of key hardware components, such as the CPU, GPU, RAM, and Storage on devices, unique fingerprints for each device are created. The performance samples are considered as time series data and used to train outlier detection transformer models, one per device and aiming to model its normal data distribution. Then, the framework is validated within a spectrum crowdsensing system leveraging Raspberry Pi devices. After a pool of experiments, the model from each device is able to individually authenticate it between the 45 devices employed for validation. An average True Positive Rate (TPR) of 0.74±0.13 and an average maximum False Positive Rate (FPR) of 0.06±0.09 demonstrate the effectiveness of this approach in enhancing authentication, security, and trust in crowdsensing applications.

In recent years, a surge of criminal activities with cross-cryptocurrency trades have emerged in Ethereum, the second-largest public blockchain platform. Most of the existing anomaly detection methods utilize the traditional machine learning with feature engineering or graph representation learning technique to capture the information in transaction network. However, these methods either ignore the timestamp information and the transaction flow direction information in transaction network or only consider single transaction network, the cross-cryptocurrency trading patterns in Ethereum are usually ignored. In this paper, we introduce a Multi-layer Temporal Transaction Anomaly Detection (MT$^2$AD) model in Ethereum network with graph neural network. Specifically, for a given Ethereum token transaction network, we first extract its initial features including the structure subgraph and edge’s feature. Then, we model the temporal information in subgraph as a series of network snapshots according to the timestamp on each edge and time window. To capture the cross-cryptocurrency trading patterns, we combine the snapshots from multiple token transactions at a given timestamp, and we consider it as a new combined graph. We further use the graph convolution encoder with attention mechanism and pooling operation on this new graph to obtain the graph-level embedding, and we transform the anomaly detection on dynamic multi-layer Ethereum transaction networks as a graph classification task with these graph-level embeddings. MT$^2$AD can integrate the transaction structure feature, edge’s feature and cross-cryptocurrency trading patterns into a framework to perform anomaly detection with graph neural networks. Experiments on three real-world multi-layer transaction networks show that the proposed MT$^2$AD (0.8789 Precision, 0.9375 Recall, 0.4987 FbMacro and 0.9351 FbWeighted) can achieve the best performance on most evaluation metrics in comparison with some competing approaches, and the effectiveness in consideration of multiple tokens is also demonstrated.