Abstract
This article proposes a feature-rich, open hardware, open software inexpensive robot based on a Waveshare AlphaBot 2. The proposal uses a Raspberry Pi and a chrome plated light bulb as a mirror to produce a robot with an omnidirectional vision (catadioptric) system. The system also tackles boot and network issues to allow for monitor-less programming and usage, thus further reducing usage costs. The OpenCV library is used for image processing and obstacles are identified based on their brightness and saturation in contrast to the ground. Our solution achieved acceptable framerates and near perfect object detection up to 1.5-m distances. The robot is usable for simple robotic demonstrations and educational purposes for its simplicity and flexibility.

Abstract
The paper provides the theoretical proof of earlier published experimental evidence of the favorable properties of a new method for State Estimation - the Generalized Correntropy Interior Point method (GCIP). The model uses a kernel estimate of the Generalized Correntropy of the error distribution as objective function, adopting Generalized Gaussian kernels. The problem is addressed by solving a constrained non-linear optimization program to maximize the similarity between states and estimated values. Solution space is searched through a special setting of a primal-dual Interior Point Method. This paper offers mathematical proof of key issues: first, that there is a theoretical shape parameter value for the kernel functions such that the feasible solution region is strictly convex, thus guaranteeing that any local solution is global or uniquely defined. Second, that a transformed system of measurement equations assures an even distribution of leverage points in the factor space of multiple regression, allowing the treatment of leverage points in a natural way. In addition, the estimated residual of GCIP model is not necessarily zero for critical (non-redundant) measurements. Finally, that the normalized residuals of critical sets are not necessarily equal in the new model, making the identification of bad data possible in these cases.

Abstract
Diabetic retinopathy (DR) grading is crucial in determining the adequate treatment and follow up of patient, but the screening process can be tiresome and prone to errors. Deep learning approaches have shown promising performance as computer-aided diagnosis (CAD) systems, but their black-box behaviour hinders clinical application. We propose DR vertical bar GRADUATE, a novel deep learning-based DR grading CAD system that supports its decision by providing a medically interpretable explanation and an estimation of how uncertain that prediction is, allowing the ophthalmologist to measure how much that decision should be trusted. We designed DR vertical bar GRADUATE taking into account the ordinal nature of the DR grading problem. A novel Gaussian-sampling approach built upon a Multiple Instance Learning framework allow DR vertical bar GRADUATE to infer an image grade associated with an explanation map and a prediction uncertainty while being trained only with image-wise labels. DR vertical bar GRADUATE was trained on the Kaggle DR detection training set and evaluated across multiple datasets. In DR grading, a quadratic-weighted Cohen's kappa (kappa) between 0.71 and 0.84 was achieved in five different datasets. We show that high kappa values occur for images with low prediction uncertainty, thus indicating that this uncertainty is a valid measure of the predictions' quality. Further, bad quality images are generally associated with higher uncertainties, showing that images not suitable for diagnosis indeed lead to less trustworthy predictions. Additionally, tests on unfamiliar medical image data types suggest that DR vertical bar GRADUATE allows outlier detection. The attention maps generally highlight regions of interest for diagnosis. These results show the great potential of DR vertical bar GRADUATE as a second-opinion system in DR severity grading.

Abstract
In wireless networking R&D we typically depend on simulation and experimentation to evaluate and validate new networking solutions. While simulations allow full control over the scenario conditions, real-world experiments are influenced by external random phenomena and may produce hardly repeatable and reproducible results, impacting the validation of the solution under evaluation. Previously, we have proposed the Trace-based Simulation (TS) approach to address the problem. TS uses traces of radio link quality and position of nodes to accurately reproduce past experiments in ns-3. Yet, in its current version, the TS approach is not compatible with scenarios where the radio spectrum is shared with concurrent networks, as it does not reproduce their channel occupancy. In this paper, we introduce the InterferencePropagationLossModel and a modified MacLow to allow reproducing the channel occupancy observed in past experiments using Wi-Fi. To validate the proposed models, the network throughput was measured in different experiments performed in the w-iLab.t testbed, controlling the channel occupancy introduced by concurrent networks. The experimental results were then compared with the network throughput achieved using the improved TS approach, the legacy TS approach, and pure simulation, validating the new proposed models and confirming their relevance to reproduce experiments previously executed in real environments. © 2020 ACM.

Abstract
Scientific publications tend to influence policymakers significantly. Despite the scientific and social importance of poverty today, the attention the top economic journals (American Economic Review; Econometrica; International Economic Review; Journal of Economic Theory; Journal of Political Economy; Quarterly Journal of Economics; Review of Economic Studies) pay to the matter is not clear, particularly in the so-called "Blue Ribbon" journals (and Review of Economics and Statistics). On the basis of bibliometric techniques, we analyzed all 27,322 articles published in the "Blue Ribbon" journals from 1970 to 2018. This is the first study on the scientific attention paid to poverty by the most influential journals in the field of economics. Two main findings can be highlighted: (i) the scientific attention paid to poverty in the Blue Ribbon journals is relatively meager, but it has observed a positive trend, increasing from a modest 0.36 percent of the total articles published in the 1970s to 1.92 percent of total publications in the 2010s; and (ii) the relative weight of specific poverty subtopics has significantly changed over the last 50 years, shifting from a focus on defining and measuring poverty in the earlier decades to policy-related issues in the most recent period (2000 onward).

Abstract
This paper introduces a synthesis procedure for the satisfiability problem of RMTL-integral formulas as SAT solving modulo theories. RMTL-integral is a real-time version of metric temporal logic (MTL) extended by a duration quantifier allowing to measure time durations. For any given formula, a SAT instance modulo the theory of arrays, uninterpreted functions with equality and non-linear real-arithmetic is synthesized and may then be further investigated using appropriate SMT solvers. We show the benefits of using RMTL-integral with the given SMT encoding on a diversified set of examples that include in particular its application in the area of schedulability analysis. Therefore, we introduce a simple language for formalizing schedulability problems and show how to formulate timing constraints as RMTL-integral formulas. Our practical evaluation based on our synthesis and Z3 as back-end SMT solver also shows the feasibility of the overall approach.