Search

The integration of Autonomous Weapon Systems (AWS) into military operations presents both significant opportunities and challenges. This paper explores the multifaceted nature of trust in AWS, emphasising the necessity of establishing reliable and transparent systems to mitigate risks associated with bias, operational failures, and accountability. Despite advancements in Artificial Intelligence (AI), the trustworthiness of these systems, especially in high-stakes military applications, remains a critical issue. Through a systematic review of existing literature, this research identifies gaps in the understanding of trust dynamics during the development and deployment phases of AWS. It advocates for a collaborative approach that includes technologists, ethicists, and military strategists to address these ongoing challenges. The findings underscore the importance of Human-Machine teaming and enhancing system intelligibility to ensure accountability and adherence to International Humanitarian Law. Ultimately, this paper aims to contribute to the ongoing discourse on the ethical implications of AWS and the imperative for trustworthy AI in defense contexts., Comment: Accepted as a workshop paper at MILCOM 2024, 8 pages

In the domain of Human-Computer Interaction, focus groups represent a widely utilised yet resource-intensive methodology, often demanding the expertise of skilled moderators and meticulous preparatory efforts. This study introduces the ``Focus Agent,'' a Large Language Model (LLM) powered framework that simulates both the focus group (for data collection) and acts as a moderator in a focus group setting with human participants. To assess the data quality derived from the Focus Agent, we ran five focus group sessions with a total of 23 human participants as well as deploying the Focus Agent to simulate these discussions with AI participants. Quantitative analysis indicates that Focus Agent can generate opinions similar to those of human participants. Furthermore, the research exposes some improvements associated with LLMs acting as moderators in focus group discussions that include human participants., Comment: 8 pages, the 24th Intelligent Virtual Agent Conference

The advent of social media transformed interpersonal communication and information consumption processes. This digital landscape accommodates user intentions, also resulting in an increase of offensive language and harmful behavior. Concurrently, social media platforms collect vast datasets comprising user-generated content and behavioral information. These datasets are instrumental for platforms deploying machine learning and data-driven strategies, facilitating customer insights and countermeasures against social manipulation mechanisms like disinformation and offensive content. Nevertheless, the availability of such datasets, along with the application of various machine learning techniques, to researchers and practitioners, for specific social media platforms regarding particular events, is limited. In particular for TikTok, which offers unique tools for personalized content creation and sharing, the existing body of knowledge would benefit from having diverse comprehensive datasets and associated data analytics solutions on offensive content. While efforts from social media platforms, research, and practitioner communities are seen on this behalf, such content continues to proliferate. This translates to an essential need to make datasets publicly available and build corresponding intelligent solutions. On this behalf, this research undertakes the collection and analysis of TikTok data containing offensive content, building a series of machine learning and deep learning models for offensive content detection. This is done aiming at answering the following research question: "How to develop a series of computational models to detect offensive content on TikTok?". To this end, a Data Science methodological approach is considered, 120.423 TikTok comments are collected, and on a balanced, binary classification approach, F1 score performance results of 0.863 is obtained., Comment: Accepted as a conference paper at DPSH 2024, 8 pages

Optically read out gaseous detectors are used in track reconstruction and imaging applications requiring high granularity images. Among resolution-determining factors, the amplification stage plays a crucial role and optimisations of detector geometry are pursued to maximise spatial resolution. To compare MicroPattern Gaseous Detector (MPGD) technologies, focused low-energy X-ray beams at the SOLEIL synchrotron facility were used to record and extract point spread function widths with Micromegas and GEM detectors. Point spread function width of $\approx$108\,\microns for Micromegas and $\approx$127\,\microns for GEM foils were extracted. The scanning of the beam with different intensities, energies and across the detector active region can be used to quantify resolution-limiting factors and improve imaging detectors using MPGD amplification stages., Comment: 22 pages, 17 figures

This paper provides a rigorous analysis of boundary element methods for the magnetic field integral equation on Lipschitz polyhedra. The magnetic field integral equation is widely used in practical applications to model electromagnetic scattering by a perfectly conducting body. The governing operator is shown to be coercive by means of the electric field integral operator with a purely imaginary wave number. Consequently, the continuous variational problem is uniquely solvable, given that the wave number does not belong to the spectrum of the interior Maxwell's problem. A Petrov-Galerkin discretization scheme is then introduced, employing Raviart-Thomas boundary elements for the solution space and Buffa-Christiansen boundary elements for the test space. Under a mild assumption depending only on the geometrical domain, the corresponding discrete inf-sup condition is proven, implying the unique solvability of the discrete problem. An asymptotically quasi-optimal error estimate for numerical solutions is established, and the convergence rate of the numerical scheme is examined. In addition, the resulting matrix system is shown to be well-conditioned regardless of the mesh refinement. Some numerical results are presented to support the theoretical analysis., Comment: 21 pages, 5 figures

Local control of superconducting circuits by high-impedance electrical gates offers potential advantages in superconducting logic, quantum processing units, and cryoelectronics. Recent experiments have reported gate-controlled supercurrent in Dayem bridges made of metallic superconductors, mediated by direct current leakage, out-of-equilibrium phonons, or possibly other mechanisms. However, a time-domain characterization of this effect has been lacking. Here, we integrate Dayem bridges made of Niobium on Silicon into coplanar-waveguide resonators, and measure the effect of the gate voltage at steady state and during pulsed operation. We consider two types of arrangements for the gate: a side-coupled gate and a remote injector. In both cases, we observe sizable changes in the real and the imaginary part of the constriction's impedance for gate voltages of the order of 1 V. However, we find striking differences in the time-domain dynamics, with the remote injector providing a faster and more controlled response. Our results contribute to our understanding of gate-controlled superconducting devices and their suitability for applications., Comment: 7 pages, 4 figures

This paper introduces a time-domain combined field integral equation for electromagnetic scattering by a perfect electric conductor. The new equation is obtained by leveraging the quasi-Helmholtz projectors, which separate both the unknown and the source fields into solenoidal and irrotational components. These two components are then appropriately rescaled to cure the solution from a loss of accuracy occurring when the time step is large. Yukawa-type integral operators of a purely imaginary wave number are also used as a Calderon preconditioner to eliminate the ill-conditioning of matrix systems. The stabilized time-domain electric and magnetic field integral equations are linearly combined in a Calderon-like fashion, then temporally discretized using an appropriate pair of trial functions, resulting in a marching-on-in-time linear system. The novel formulation is immune to spurious resonances, dense discretization breakdown, large-time step breakdown and dc instabilities stemming from non-trivial kernels. Numerical results for both simply-connected and multiply-connected scatterers corroborate the theoretical analysis., Comment: 13 pages

This paper aims to address two issues of integral equations for the scattering of time-harmonic electromagnetic waves by a perfect electric conductor with Lipschitz continuous boundary: ill-conditioned {boundary element Galerkin matrices} on fine meshes and instability at spurious resonant frequencies. The remedy to ill-conditioned matrices is operator preconditioning, and resonant instability is eliminated by means of a combined field integral equation. Exterior traces of single and double layer potentials are complemented by their interior counterparts for a purely imaginary wave number. We derive the corresponding variational formulation in the natural trace space for electromagnetic fields and establish its well-posedness for all wave numbers. A Galerkin discretization scheme is employed using conforming edge boundary elements on dual meshes, which produces well-conditioned discrete linear systems of the variational formulation. Some numerical results are also provided to support the numerical analysis., Comment: 6 figures

Macroscopic superconducting components are an important building block of various quantum circuits. Since several of the envisioned applications require exposure to magnetic fields, it is of utmost importance to explore the impact of magnetic fields on their performance. Here we explore the complex pattern of magnetic field penetration and identify its impact on the resonance frequency of NbTiN superconducting resonators by combining magneto-optical imaging and high-frequency measurements. At temperatures below approximately half of the superconducting critical temperature, the development of magnetic flux avalanches manifests itself as a noisy response in the field-dependent resonance frequency. Magneto-optical imaging reveals different regimes and distinguishes the impact of avalanches in the ground plane and resonator. Our findings demonstrate that superconducting resonators represent a valuable tool to investigate magnetic flux dynamics. Moreover, the current blooming of niobium-based superconducting radio-frequency devices makes this report timely by unveiling the severe implications of magnetic flux dynamics., Comment: 20 pages, 3 figures, 2 supp. figures, 3 supp. videos

In the last years, optical readout of Micromegas gaseous detectors has been achieved by implementing a Micromegas detector on a glass anode coupled to a CMOS camera. Effective X-ray radiography was demonstrated using integrated imaging approach. High granularity values have been reached for low-energy X-rays from radioactive sources and X-ray generators. Detector characterization with X-ray radiography has led to two applications: neutron imaging for non-destructive examination of highly gamma-ray emitting objects and beta imaging for the single cell activity tagging in the field of oncology drug studies. First measurements investigating the achievable spatial resolution of the glass Micromegas detector at the SOLEIL synchrotron facility with a high-intensity and flat irradiation field will be shown in this article., Comment: 6 pages, 4 figures, 7th International Conference on Micro Pattern Gaseous Detectors, 11-16 December 20223, Weizmann Institute of Science, Rehovot, Israel

Different factors have effects on traffic crashes and crash-related injuries. These factors include segment characteristics, crash-level characteristics, occupant level characteristics, environment characteristics, and vehicle level characteristics. There are several studies regarding these factors' effects on crash injuries. However, limited studies have examined the effects of pre-crash events on injuries, especially for curve-related crashes. The majority of previous studies for curve-related crashes focused on the impact of geometric features or street design factors. The current study tries to eliminate the aforementioned shortcomings by considering important pre-crash events related factors as selected variables and the number of vehicles with or without injury as the predicted variable. This research used CRSS data from the National Highway Traffic Safety Administration (NHTSA), which includes traffic crash-related data for different states in the USA. The relationships are explored using different machine learning algorithms like the random forest, C5.0, CHAID, Bayesian Network, Neural Network, C\&R Tree, Quest, etc. The random forest and SHAP values are used to identify the most effective variables. The C5.0 algorithm, which has the highest accuracy rate among the other algorithms, is used to develop the final model. Analysis results revealed that the extent of the damage, critical pre-crash event, pre-impact location, the trafficway description, roadway surface condition, the month of the crash, the first harmful event, number of motor vehicles, attempted avoidance maneuver, and roadway grade affect the number of vehicles with or without injury in curve-related crashes.

We propose a preconditioner for the Helmholtz exterior problems on multi-screens. For this, we combine quotient-space BEM and operator preconditioning. For a class of multi-screens (which we dub \emph{type A} multi-screens), we show that this approach leads to block diagonal Calder\'on preconditioners and results in a spectral condition number that grows only logarithmically with $h$, just as in the case of simple screens. Since the resulting scheme contains many more DoFs than strictly required, we also present strategies to remove almost all redundancy without significant loss of effectiveness of the preconditioner. We verify these findings by providing representative numerical results. Further numerical experiments suggest that these results can be extended beyond type A multi-screens and that the numerical method introduced here can be applied to essentially all multi-screens encountered by the practitioner, leading to a significantly reduced simulation cost.

Machine-learning based classifiers have become indispensable in the field of astrophysics, allowing separation of astronomical sources into various classes, with computational efficiency suitable for application to the enormous data volumes that wide-area surveys now typically produce. In the standard supervised classification paradigm, a model is typically trained and validated using data from relatively small areas of sky, before being used to classify sources in other areas of the sky. However, population shifts between the training examples and the sources to be classified can lead to `silent' degradation in model performance, which can be challenging to identify when the ground-truth is not available. In this Letter, we present a novel methodology using the NannyML Confidence-Based Performance Estimation (CBPE) method to predict classifier F1-score in the presence of population shifts, but without ground-truth labels. We apply CBPE to the selection of quasars with decision-tree ensemble models, using broad-band photometry, and show that the F1-scores are predicted remarkably well (MAPE ~ 10%; R^2 = 0.74-0.92). We discuss potential use-cases in the domain of astronomy, including machine-learning model and/or hyperparameter selection, and evaluation of the suitability of training datasets for a particular classification problem., Comment: 5.5 pages, 2 figures, accepted for publication in MNRAS Letters