Search

The MOOSE (Monitoring Observations of SMC X-1 Excursions) program uses the Neutron Star Interior Composition Explorer Mission (NICER) to monitor the high mass X-ray binary SMC X-1 during its superorbital period excursions. Here we perform X-ray spectral analyses of 26 NICER observations of SMC X-1, taken at the tail-end of the excursion between 2021-04-01 and 2022-01-05. We use a single spectral model to fit spectra observed in high, intermediate and low states, using a combination of a partial covering fraction model, a black-body disc, and a power-law component. We find that the partial covering fraction varies significantly with the superorbital state during superorbital excursion. Our findings suggest that the low/high state in SMC X-1 is caused by a very high obscuration of the accretion disk., Comment: 10 pages 4 figures, accepted to MNRAS

Ontologies and knowledge graphs (KGs) are general-purpose computable representations of some domain, such as human anatomy, and are frequently a crucial part of modern information systems. Most of these structures change over time, incorporating new knowledge or information that was previously missing. Managing these changes is a challenge, both in terms of communicating changes to users, and providing mechanisms to make it easier for multiple stakeholders to contribute. To fill that need, we have created KGCL, the Knowledge Graph Change Language, a standard data model for describing changes to KGs and ontologies at a high level, and an accompanying human-readable controlled natural language. This language serves two purposes: a curator can use it to request desired changes, and it can also be used to describe changes that have already happened, corresponding to the concepts of "apply patch" and "diff" commonly used for managing changes in text documents and computer programs. Another key feature of KGCL is that descriptions are at a high enough level to be useful and understood by a variety of stakeholders--for example, ontology edits can be specified by commands like "add synonym 'arm' to 'forelimb'" or "move 'Parkinson disease' under 'neurodegenerative disease'". We have also built a suite of tools for managing ontology changes. These include an automated agent that integrates with and monitors GitHub ontology repositories and applies any requested changes, and a new component in the BioPortal ontology resource that allows users to make change requests directly from within the BioPortal user interface. Overall, the KGCL data model, its controlled natural language, and associated tooling allow for easier management and processing of changes associated with the development of ontologies and KGs.

We simulate mergers between star clusters embedded within their natal giant molecular cloud. We extract initial conditions from cloud-scale simulations of cluster formation and introduce different prescriptions for primordial binaries. We find that simulations that do not include primordial binaries result in a larger fraction of unbound stars than simulations which include a prescription for binaries based on observations. We also find a preferred direction of motion for stars that become unbound during the merger. Sub-cluster mergers within realistic gas environments promote binary disruption while mergers between idealized, gas-rich spherical clusters do not produce the same disruption. Binary systems with smaller semi-major axes are disrupted in simulations of sub-cluster mergers within their natal environment compared to simulations that do not include the realistic gas environment. We conclude that binary disruption and the production of an anisotropic distribution of unbound stars are the natural consequences of sub-cluster mergers during star cluster assembly., Comment: 13 pages, 8 figures. Re-submitted to ApJ following favourable referee report

Climate change will result in more frequent, more intense and longer-lasting heat waves by 2050. As part of its Climate Plan and its resilience strategy, the City of Paris is deploying, through its Oasis program, a network of urban cool islands to mitigate the urban heat island phenomena: schoolyards are renovated in order to reduce the heat stress of users. We establish a methodology aiming to quantify the microclimatic impact of the transformation. Mobile measurements are carried out within a case courtyard under hot conditions and coupled with fixed weather station data to evaluate heat stress using UTCI. The heat stress mapping thus obtained allows a first microclimatic diagnosis of the schoolyard., Comment: in French language, 34e Colloque Annuel de l'Association Internationale de Climatologie, Jul 2021, Casablanca, Maroc

We prove that for every integer $d \ge 2$ there exists a dense collection of subsets of $[n]^d$ such that no two of them have a symmetric difference that may be written as the $d$th power of a union of at most $\lfloor d/2 \rfloor$ intervals. This provides a limitation on reasonable tightenings of a question of Alon from 2023 and of a conjecture of Gowers from 2009, and investigates a direction analogous to that of recent works of Conlon, Kam\v{c}ev, Leader, R\"aty and Spiegel on intervals in the Hales-Jewett theorem., Comment: 13 pages, submitted

This research paper introduces a novel word-level Optical Character Recognition (OCR) model specifically designed for digital Urdu text, leveraging transformer-based architectures and attention mechanisms to address the distinct challenges of Urdu script recognition, including its diverse text styles, fonts, and variations. The model employs a permuted autoregressive sequence (PARSeq) architecture, which enhances its performance by enabling context-aware inference and iterative refinement through the training of multiple token permutations. This method allows the model to adeptly manage character reordering and overlapping characters, commonly encountered in Urdu script. Trained on a dataset comprising approximately 160,000 Urdu text images, the model demonstrates a high level of accuracy in capturing the intricacies of Urdu script, achieving a CER of 0.178. Despite ongoing challenges in handling certain text variations, the model exhibits superior accuracy and effectiveness in practical applications. Future work will focus on refining the model through advanced data augmentation techniques and the integration of context-aware language models to further enhance its performance and robustness in Urdu text recognition.

We show that a conceptually simple covering technique has surprisingly rich applications to density theorems and conjectures on patterns in sets involving set differences. These applications fall into three categories: (i) analogues of these statements to distance $2$ versions of the pattern, (ii) reduction of these statements to relative versions, and (iii) reductions of these statements to a quasirandom case with respect to some quantities that affect the number of realisations of the pattern., Comment: 25 pages, submitted

Paris is experimenting and implementing strategies to increase the capital's resilience and promote climate change adaptation. Major heatwaves have been hitting the French Capital lately (Bador et al., 2017) and have thus focused attention on UHI countermeasures, such as pavement-watering solutions and urban greening. The Oasis Project is one such strategy aiming to transform schoolyards into urban cool islands that would benefit surrounding neighborhoods and their inhabitants during heatwaves.The work presented here focuses on identifying high-priority schoolyards among the 670 city-owned schools. This is conducted using a GIS tool used to identify areas with high cooling potential, which would benefit most from UHI countermeasures. After extracting bare schoolyard from the facilities, we built a cooling indicator based on the ratio of high solar irradiance surface to the whole schoolyard area. We were thus able to identify 38 schoolyards with high cooling potential, 157 with medium cooling potential and 286 facilities with moderate cooling potential, out of the 670 facilities.The methodology can be applied to other cities, and therefore helps set up GIS tools that can provide municipalities with meaningful insight into their urban cooling strategy.

As part of its Resilience Strategy, the City of Paris' OASIS program aims to contribute ot its adaptation to heatwaves and climate change by transforming schoolyards into climate shelters, namely via desealing and greening. In this context, a variety of alternative pavement materials have been proposed to replace the initial schoolyard pavement, composed of an asphalt sidewalk structure. In the context of the EU-funded ERDF UIA OASIS project, the thermal performance of these alternative materials and their impact in terms of urban cooling was explored. To this aim, five samples of reference and innovative schoolyard pavements were studied in the lab under heat-wave conditions. Alternative green, biosourced, recycled and reflective pavement solutions were compared to standard fine-aggregate asphalt concrete. Their performance was evaluated with regards to their contribution to the urban heat island phenomenon and to pedestrian heat stress, account for the typical use schedule of schoolyards. Green and biosourced materials were found to perform well for both indicators, while the standard and recycled solutions had poor UHI performance but had limited negative effects on daytime heat stress. The reflective pavement had better UHI performance but had high radiosity during daytime which can negatively affect pedestrian heat stress.

In the aftermath of the 2003 heatwave, and with growing concern over climate change, Paris City Hall has been implementing several heat mitigation strategies. One of these is the OASIS Schoolyard Strategy which aims to transform Parisian schoolyards into cool islands. Within this framework, the EU-funded ERDF UIA OASIS Project aims to study the transformation of ten schoolyards, including an evaluation of their microclimatic performance.The present article presents case study results from one schoolyard using GIS data and fixed and mobile microclimatic measurements. An analysis method for mobile measurement data is proposed, tested and discussed on the basis of this case study.

Infrared imagery can help in low-visibility situations such as fog and low-light scenarios, but it is prone to thermal noise and requires further processing and correction. This work studies the effect of different infrared processing pipelines on the performance of a pedestrian detection in an urban environment, similar to autonomous driving scenarios. Detection on infrared images is shown to outperform that on visible images, but the infrared correction pipeline is crucial since the models cannot extract information from raw infrared images. Two thermal correction pipelines are studied, the shutter and the shutterless pipes. Experiments show that some correction algorithms like spatial denoising are detrimental to performance even if they increase visual quality for a human observer. Other algorithms like destriping and, to a lesser extent, temporal denoising, increase computational time, but have some role to play in increasing detection accuracy. As it stands, the optimal trade-off for speed and accuracy is simply to use the shutterless pipe with a tonemapping algorithm only, for autonomous driving applications within varied environments., Comment: 9 pages, 7 figures, 4 tables

Meta-Reinforcement Learning (Meta-RL) aims to acquire meta-knowledge for quick adaptation to diverse tasks. However, applying these policies in real-world environments presents a significant challenge in balancing rapid adaptability with adherence to environmental constraints. Our novel approach, Constraint Model Agnostic Meta Learning (C-MAML), merges meta learning with constrained optimization to address this challenge. C-MAML enables rapid and efficient task adaptation by incorporating task-specific constraints directly into its meta-algorithm framework during the training phase. This fusion results in safer initial parameters for learning new tasks. We demonstrate the effectiveness of C-MAML in simulated locomotion with wheeled robot tasks of varying complexity, highlighting its practicality and robustness in dynamic environments.

We recognise that an entropy inequality akin to the main intermediate goal of recent works (Gowers, Green, Manners, Tao [3],[2]) regarding a conjecture of Marton provides a black box from which we can also through a short deduction recover another description: if a finite subset $A$ of an abelian group $G$ is such that the distribution of the sums $a+b$ with $(a,b) \in A \times A$ is only slightly more spread out than the uniform distribution on $A$, then $A$ has small symmetric difference with some finite coset of $G$. The resulting bounds are necessarily sharp up to a logarithmic factor., Comment: 10 pages, submitted

After the repeal of Roe vs. Wade in June 2022, women face long-distance travel across state lines to access abortion care. For women who also face socioeconomic hardship, travel for abortion care is a significant burden. To ease this burden, abortion access nonprofits are funding and/or supplying transportation to abortion clinics. However, due to the uneven distribution of demand and supply for abortions, these nonprofits do not have efficient logistical operations. As a result, low-income, underserved women may not have access to adequate reproductive healthcare, thus widening healthcare inequity gaps. Nonprofits may also risk not serving the needs of vulnerable women without access to adequate reproductive healthcare, and in doing so, waste resources, money, and volunteer hours. To address these challenges, we create an interactive, web-based planning tool, the Reproductive Healthcare Equity Algorithm (RHEA), to guide nonprofits in strategically allocating resources and serving demand. RHEA leverages an optimization model to determine the maximum flow and minimum transportation cost to route women across a network of counties and abortion clinics, subject to transportation supply, budget, and time constraints for one day of operations for a nonprofit. In doing so, we collaborate with abortion access nonprofits to cater our model design and interface development to their needs and considerations. Ultimately, we seek to optimize resource allocation for nonprofits providing abortion care logistics and improve abortion access for low-income, underserved women.

On the forefront of scientific computing, Deep Learning (DL), i.e., machine learning with Deep Neural Networks (DNNs), has emerged a powerful new tool for solving Partial Differential Equations (PDEs). It has been observed that DNNs are particularly well suited to weakening the effect of the curse of dimensionality, a term coined by Richard E. Bellman in the late `50s to describe challenges such as the exponential dependence of the sample complexity, i.e., the number of samples required to solve an approximation problem, on the dimension of the ambient space. However, although DNNs have been used to solve PDEs since the `90s, the literature underpinning their mathematical efficiency in terms of numerical analysis (i.e., stability, accuracy, and sample complexity), is only recently beginning to emerge. In this paper, we leverage recent advancements in function approximation using sparsity-based techniques and random sampling to develop and analyze an efficient high-dimensional PDE solver based on DL. We show, both theoretically and numerically, that it can compete with a novel stable and accurate compressive spectral collocation method. In particular, we demonstrate a new practical existence theorem, which establishes the existence of a class of trainable DNNs with suitable bounds on the network architecture and a sufficient condition on the sample complexity, with logarithmic or, at worst, linear scaling in dimension, such that the resulting networks stably and accurately approximate a diffusion-reaction PDE with high probability.

More than ever, nations around the world understand that science, technology, engineering, and mathematics (STEM) skills are key to driving economic growth and overall competitiveness. A compact between the United States, through the Millennium Challenge Corporation, and the country of Georgia's government sought to bring international higher education to Georgians -- particularly in the area of STEM -- to develop human capital, advance economic growth, and reduce poverty in the country. This effort included a project to ensure and enhance the delivery of high-quality STEM bachelor's degrees in Georgia in four ways: (1) partnering with a U.S. university to offer U.S. STEM degree programs, (2) improving STEM-related infrastructure and technology in partner Georgian universities, (3) assisting partner Georgian universities in reaching international STEM standards and acquiring international program accreditation, and (4) preparing world-class skilled STEM specialists from diverse backgrounds. San Diego State University (SDSU) partnered with three Georgian public universities -- Tbilisi State University, Georgian Technical University, and Ilia State University -- to fulfill these goals. In this report, the authors assess the project's implementation, graduate outcomes, sustainability, and return on investment.

In wireless communication Multiple Input Multiple Output (MIMO) technology has brought significant improvement in service by adopting Orthogonal Frequency Division Multiplexing (OFDM), a digital modulation technique. To achieve great performance with MIMO efficiently gathering channel state information (CSI) plays a vital role. Among different approach of channel estimation techniques data-aided channel estimation is more reliable. The existing methods of data-aided channel estimation are Least Square (LS) and Minimum Mean Square Error (MMSE) methods which do not achieve a great performance. Moreover, MMSE is little complex and has higher computational cost. That is why many attempts have been done previously to optimize the methods with help of meta heuristics and also other ways. In this paper we have tried to optimize LS estimation with a combined algorithm of Genetic Algorithm (GA) and Particle Swarm Optimization (PSO). The proposed algorithm has outperformed LS and MMSE. And it gives similar result if we optimize LS with standard PSO but in less numbers of iteration.

We identify an assumption on linear forms $\phi_1, \dots, \phi_k: \mathbb{F}_p^n \to \mathbb{F}_p$ that is much weaker than approximate joint equidistribution on the Boolean cube $\{0,1\}^n$ and is in a sense almost as weak as linear independence, but which guarantees that every subset of $\{0,1\}^n$ on which none of $\phi_1, \dots, \phi_k$ has full image has a density which tends to 0 with $k$. This density is at most quasipolynomially small in $k$, a bound that is necessarily close to sharp., Comment: 17 pages, submitted version

FAIR data presupposes their successful communication between machines and humans while preserving their meaning and reference, requiring all parties involved to share the same background knowledge. Inspired by English as a natural language, we investigate the linguistic structure that ensures reliable communication of information and draw parallels with data structures, understanding both as models of systems of interest. We conceptualize semantic interoperability as comprising terminological and propositional interoperability. The former includes ontological (i.e., same meaning) and referential (i.e., same referent/extension) interoperability and the latter schema (i.e., same data schema) and logical (i.e., same logical framework) interoperability. Since no best ontology and no best data schema exists, establishing semantic interoperability and FAIRness of data and metadata requires the provision of a comprehensive set of relevant ontological and referential entity mappings and schema crosswalks. We therefore propose appropriate additions to the FAIR Guiding Principles, leading to FAIR 2.0. Furthermore, achieving FAIRness of data requires the provision of FAIR services in addition to organizing data into FAIR Digital Objects. FAIR services include a terminology, a schema, and an operations service.

Machine learning (ML) techniques have been applied to high-level synthesis (HLS) flows for quality-of-result (QoR) prediction and design space exploration (DSE). Nevertheless, the scarcity of accessible high-quality HLS datasets and the complexity of building such datasets present challenges. Existing datasets have limitations in terms of benchmark coverage, design space enumeration, vendor extensibility, or lack of reproducible and extensible software for dataset construction. Many works also lack user-friendly ways to add more designs, limiting wider adoption of such datasets. In response to these challenges, we introduce HLSFactory, a comprehensive framework designed to facilitate the curation and generation of high-quality HLS design datasets. HLSFactory has three main stages: 1) a design space expansion stage to elaborate single HLS designs into large design spaces using various optimization directives across multiple vendor tools, 2) a design synthesis stage to execute HLS and FPGA tool flows concurrently across designs, and 3) a data aggregation stage for extracting standardized data into packaged datasets for ML usage. This tripartite architecture ensures broad design space coverage via design space expansion and supports multiple vendor tools. Users can contribute to each stage with their own HLS designs and synthesis results and extend the framework itself with custom frontends and tool flows. We also include an initial set of built-in designs from common HLS benchmarks curated open-source HLS designs. We showcase the versatility and multi-functionality of our framework through six case studies: I) Design space sampling; II) Fine-grained parallelism backend speedup; III) Targeting Intel's HLS flow; IV) Adding new auxiliary designs; V) Integrating published HLS data; VI) HLS tool version regression benchmarking. Code at https://github.com/sharc-lab/HLSFactory., Comment: Edit to "Section V.E" for proper attribution of open-source HLSyn, AutoDSE, and the Merlin compiler

This work will study an optimal control problem describing the two-strain SEIR epidemic model. The studied model is in the form of six nonlinear differential equations illustrating the dynamics of the susceptibles and the exposed, the infected, and the recovered individuals. The exposed and the infected compartments are each divided into two sub-classes representing the first and the second strain. The model includes two saturated rates and two treatments for each strain. We begin our study by showing the well-posedness of our problem. The basic reproduction number is calculated and depends mainly on the reproduction numbers of the first and second strains. The global stability of the disease-free equilibrium is fulfilled. The optimal control study is achieved by using the Pontryagin minimum principle. Numerical simulations have shown the importance of therapy in minimizing the infection's effect. By administrating suitable therapies, the disease's severity decreases considerably. The estimation of parameters as well as a comparison study with COVID-19 clinical data are fulfilled. It was shown that the mathematical model results fits well the clinical data.

A quantitative analysis of post-VVC luma and chroma intra tools is presented, focusing on their statistical behaviors, in terms of block selection rate under different conditions. The aim is to provide insights to the standardization community, offering a clearer understanding of interactions between tools and assisting in the design of an optimal combination of these novel tools when the JVET enters the standardization phase. Specifically, this paper examines the selection rate of intra tools as function of 1) the version of the ECM, 2) video resolution, and 3) video bitrate. Additionally, tests have been conducted on sequences beyond the JVET CTC database. The statistics show several trends and interactions, with various strength, between coding tools of both luma and chroma., Comment: Submitted to IEEE ICIP 2024

We perform high resolution simulations of forming star clusters as they merge inside giant molecular clouds (GMCs) using hydrodynamics coupled to N-body dynamics to simultaneously model both the gas and stars. We zoom in to previously run GMC simulations and resolve clusters into their stellar and gas components while including the surrounding GMC environment. We find that GMC gas is important in facilitating the growth of clusters in their embedded phase by promoting cluster mergers. Mergers induce asymmetric expansion of the stellar component of the clusters in our simulations. As well, mergers induce angular momentum in the clusters' stellar and gas components. We find that mergers can lead to an increase in the amount of dense gas present in clusters if a background gas distribution is present. We predict that this can lead to new star formation that can change the overall distribution of cluster stars in velocity space. Our results suggest that subcluster mergers in the presence of background gas can imprint dynamical signatures that can be used to constrain cluster formation histories., Comment: Accepted for publication in The Astrophysical Journal, 15 pages, 10 figures

As Artificial Intelligence (AI) making advancements in medical decision-making, there is a growing need to ensure doctors develop appropriate reliance on AI to avoid adverse outcomes. However, existing methods in enabling appropriate AI reliance might encounter challenges while being applied in the medical domain. With this regard, this work employs and provides the validation of an alternative approach -- majority voting -- to facilitate appropriate reliance on AI in medical decision-making. This is achieved by a multi-institutional user study involving 32 medical professionals with various backgrounds, focusing on the pathology task of visually detecting a pattern, mitoses, in tumor images. Here, the majority voting process was conducted by synthesizing decisions under AI assistance from a group of pathology doctors (pathologists). Two metrics were used to evaluate the appropriateness of AI reliance: Relative AI Reliance (RAIR) and Relative Self-Reliance (RSR). Results showed that even with groups of three pathologists, majority-voted decisions significantly increased both RAIR and RSR -- by approximately 9% and 31%, respectively -- compared to decisions made by one pathologist collaborating with AI. This increased appropriateness resulted in better precision and recall in the detection of mitoses. While our study is centered on pathology, we believe these insights can be extended to general high-stakes decision-making processes involving similar visual tasks., Comment: 46 pages, 11 figures. Accepted International Journal of Human-Computer Studies

In the literature, machine learning (ML) has been implemented at the base station (BS) and user equipment (UE) to improve the precision of downlink channel state information (CSI). However, ML implementation at the UE can be infeasible for various reasons, such as UE power consumption. Motivated by this issue, we propose a CSI learning mechanism at BS, called CSILaBS, to avoid ML at UE. To this end, by exploiting channel predictor (CP) at BS, a light-weight predictor function (PF) is considered for feedback evaluation at the UE. CSILaBS reduces over-the-air feedback overhead, improves CSI quality, and lowers the computation cost of UE. Besides, in a multiuser environment, we propose various mechanisms to select the feedback by exploiting PF while aiming to improve CSI accuracy. We also address various ML-based CPs, such as NeuralProphet (NP), an ML-inspired statistical algorithm. Furthermore, inspired to use a statistical model and ML together, we propose a novel hybrid framework composed of a recurrent neural network and NP, which yields better prediction accuracy than individual models. The performance of CSILaBS is evaluated through an empirical dataset recorded at Nokia Bell-Labs. The outcomes show that ML elimination at UE can retain performance gains, for example, precoding quality., Comment: 14 pages, 11 figures

This study investigates the emission of cations from silica samples by single-cycle THz pulses, focusing on the influence of pulse polarity. Negative THz pulses were found to efficiently trigger the evaporation of cations from nanoneedles in amorphous silica samples compared to positive pulses. Conversely, this dependence on pulse polarity could not be found in samples with metallic behaviour such as LaB$_6$ and when multi-cycle pulses in different frequency ranges such as ultraviolet (UV) are used. First-principles simulations focus on silica under THz laser irradiation and show critical fields for ion evaporation of hydroxyl groups from Si(OH)$_4$, which serves as a model precursor molecule for the amorphous solid matrix. To explain our experimental results, we propose a simplified theoretical model that determines the role of the polarity of the THz pulse by taking into account the differences in electron mobility between silica and semi-metallic samples. The study explores the nonlinear microscopic mechanisms of atomic evaporation under external static and THz laser fields and clarifies the dynamics of THz-enhanced APT and related applications.