Search

Tungsten-based low-activation high-entropy alloys are possible candidates for next-generation fusion reactors due to their exceptional tolerance to irradiation, thermal loads, and stress. We develop an accurate and efficient machine-learned interatomic potential for the W-Ta-Cr-V system and use it in hybrid Monte Carlo molecular dynamics simulations of ordering and segregation to all common types of defects in WTaCrV. The predictions are compared to atom probe tomography analysis of segregation and precipitation in WTaCrV thin films. By also considering two other alloys, WTaV and MoNbTaVW, we are able to draw general conclusions about preferred segregation in refractory alloys and the reasons behind it, guiding future alloy design and elucidating experimental observations. We show that the experimentally observed CrV precipitates in WTaCrV form semicoherent bcc-to-bcc interfaces with the surrounding matrix, as coherent precipitates are not thermodynamically stable due to excessive lattice mismatch. The predictions from simulations align well with our atom probe tomography analysis as well as previous experimental observations.

Refractory High-Entropy Alloys (RHEAs) hold promising potential to be used as structural materials in future nuclear fusion reactors, where W and its alloys are currently leading candidates. Fusion materials must be able to withstand extreme conditions, such as (i) severe radiation-damage arising from highly-energetic neutrons, (ii) embrittlement caused by implantation of H and He ions, and (iii) exposure to extreme high-temperatures and thermal gradients. Recent research demonstrated that two RHEAs - the WTaCrV and WTaCrVHf - can outperform both coarse-grained and nanocrystalline W in terms of its radiation response and microstructural stability. Chemical complexity and nanocrystallinity enhance the radiation tolerance of these new RHEAs, but their multi-element nature, including low-melting Cr, complicates bulk fabrication and limits practical applications. We demonstrate that reducing the number of alloying elements and yet retain high-radiation tolerance is possible within the ternary system W-Ta-V via synthesis of two novel nanocrystalline refractory medium-entropy alloys (RMEAs): the W$_{53}$Ta$_{44}$V$_{3}$ and W$_{53}$Ta$_{42}$V$_{5}$ (in at.\%). We experimentally show that the radiation response of the W-Ta-V system can be tailored by small additions of V, and such experimental result was validated with theoretical analysis of chemical short-range orders (CSRO) from combined ab-initio atomistic Monte-Carlo modeling. It is predicted from computational analysis that a small change in V concentration has a significant effect on the Ta-V CRSO between W$_{53}$Ta$_{44}$V$_{3}$ and W$_{53}$Ta$_{42}$V$_{5}$ leading to radiation-resistant microstructures in these RMEAs from chemistry stand-point of views. We deviate from the original high-entropy alloy concept to show that high radiation resistance can be achieved in systems with simplified chemical complexity.

We introduce a neuro-symbolic transformer-based model that converts flat, segmented facade structures into procedural definitions using a custom-designed split grammar. To facilitate this, we first develop a semi-complex split grammar tailored for architectural facades and then generate a dataset comprising of facades alongside their corresponding procedural representations. This dataset is used to train our transformer model to convert segmented, flat facades into the procedural language of our grammar. During inference, the model applies this learned transformation to new facade segmentations, providing a procedural representation that users can adjust to generate varied facade designs. This method not only automates the conversion of static facade images into dynamic, editable procedural formats but also enhances the design flexibility, allowing for easy modifications., Comment: 11 pages, 11 figures, in ACM SIGGRAPH Asia 2024 Conference Papers Proceedings

In this paper, we investigate the problem of strong approximation of the solution of SDEs in the case when the drift coefficient is given in the integral form. Such drift often appears when analyzing stochastic dynamics of optimization procedures in machine learning problems. We discuss connections of the defined randomized Euler approximation scheme with the perturbed version of the stochastic gradient descent (SGD) algorithm. We investigate its upper error bounds, in terms of the discretization parameter n and the size M of the random sample drawn at each step of the algorithm, in different subclasses of coefficients of the underlying SDE. Finally, the results of numerical experiments performed by using GPU architecture are also reported.

The Omnibus Directive is an essential part of the European Union's New Deal for Consumers. The Directive introduces new regulations in trade, including e-commerce, with the main goal being to increase transparency, fairness and consumer protection. The authors critically draw attention to a significant oversight in the Omnibus Directive, namely the lack of consideration of recommendation systems. Recommendation engines can be a source of potentially harmful practices affecting consumers, hence the need for a directive extension. The proposals presented in this article include the introduction of ethical supervision over recommendation systems to minimize the risk of negative effects of their recommendations, as well as a clear explanation of the criteria on which recommendations are made -- similar to search result rankings. -- Dyrektywa Omnibus stanowi istotn\k{a} cz\k{e}\'s\'c Nowego {\L}adu dla Konsument\'ow (ang. \emph{New Deal for Consumers}) Unii Europejskiej. Dyrektywa wprowadza nowe regulacje w handlu, w tym e-commerce, kt\'orych g{\l}\'ownym celem jest zwi\k{e}kszenie przejrzysto\'sci, uczciwo\'sci i ochrony konsument\'ow. Autorzy krytycznie zwracaj\k{a} uwag\k{e} na istotne zaniedbanie w dyrektywie Omnibus, jakim jest brak uwzgl\k{e}dnienia system\'ow rekomendacyjnych. Silniki rekomendacyjne mog\k{a} by\'c \'zr\'od{\l}em potencjalnie szkodliwych praktyk uderzaj\k{a}cych w konsument\'ow, st\k{a}d niezb\k{e}dne jest rozszerzenie dyrektywy. Propozycje przedstawione w niniejszym artykule obejmuj\k{a} wprowadzenie etycznego nadzoru nad systemami rekomenduj\k{a}cymi, aby zminimalizowa\'c ryzyko negatywnych skutk\'ow ich rekomendacji, a tak\.ze jasne wyja\'snienie kryteri\'ow, na podstawie kt\'orych dokonywane s\k{a} rekomendacje -- analogicznie do ranking\'ow wynik\'ow wyszukiwania., Comment: 7 pages, in Polish

In this paper, we investigate the properties of standard and multilevel Monte Carlo methods for weak approximation of solutions of stochastic differential equations (SDEs) driven by the infinite-dimensional Wiener process and Poisson random measure with Lipschitz payoff function. The error of the truncated dimension randomized numerical scheme, which is determined by two parameters, i.e grid density $n \in \mathbb{N}_{+}$ and truncation dimension parameter $M \in \mathbb{N}_{+},$ is of the order $n^{-1/2}+\delta(M)$ such that $\delta(\cdot)$ is positive and decreasing to $0$. We derive complexity model and provide proof for the upper complexity bound of the multilevel Monte Carlo method which depends on two increasing sequences of parameters for both $n$ and $M.$ The complexity is measured in terms of upper bound for mean-squared error and compared with the complexity of the standard Monte Carlo algorithm. The results from numerical experiments as well as Python and CUDA C implementation are also reported., Comment: 23 pages, 4 figures, 2 code listings

This paper focuses on analyzing the error of the randomized Euler algorithm when only noisy information about the coefficients of the underlying stochastic differential equation (SDE) and the driving Wiener process is available. Two classes of disturbed Wiener process are considered, and the dependence of the algorithm's error on the regularity of the disturbing functions is investigated. The paper also presents results from numerical experiments to support the theoretical findings.

We propose a novel methodology for estimating the epidemiological parameters of a modified SIRD model (acronym of Susceptible, Infected, Recovered and Deceased individuals) and perform a short-term forecast of SARS-CoV-2 virus spread. We mainly focus on forecasting number of deceased. The procedure was tested on reported data for Poland. For some short-time intervals we performed numerical test investigating stability of parameter estimates in the proposed approach. Numerical experiments confirm the effectiveness of short-term forecasts (up to 2 weeks) and stability of the method. To improve their performance (i.e. computation time) GPU architecture was used in computations., Comment: 17 pages, 7 figures

In this paper we deal with pointwise approximation of solutions of stochastic differential equations (SDEs) driven by infinite dimensional Wiener process with additional jumps generated by Poisson random measure. The further investigations contain upper error bounds for the proposed truncated dimension randomized Euler scheme. We also establish matching (up to constants) upper and lower bounds for $\varepsilon$-complexity and show that the defined algorithm is optimal in the Information-Based Complexity (IBC) sense. Finally, results of numerical experiments performed by using GPU architecture are also reported.

This paper proposes sound event localization and detection methods from multichannel recording. The proposed system is based on two Convolutional Recurrent Neural Networks (CRNNs) to perform sound event detection (SED) and time difference of arrival (TDOA) estimation on each pair of microphones in a microphone array. In this paper, the system is evaluated with a four-microphone array, and thus combines the results from six pairs of microphones to provide a final classification and a 3-D direction of arrival (DOA) estimate. Results demonstrate that the proposed approach outperforms the DCASE 2019 baseline system.

A novel W-based refractory high entropy alloy with outstanding radiation resistance has been developed. The alloy was grown as thin films showing a bimodal grain size distribution in the nanocrystalline and ultrafine regimes and a unique 4 nm lamella-like structure revealed by atom probe tomography (APT). Transmission electron microscopy (TEM) and X-ray diffraction show an underlying body-centered cubic crystalline structure with certain black spots appearing after thermal annealing at elevated temperatures. Thorough analysis based on TEM and APT correlated the black spots with second phase particles rich in Cr and V. After both in situ and ex situ irradiation, these precipitates evolve to quasi-spherical particles with no sign of irradiation-created dislocation loops even after 8 dpa at either room temperature or 1073 K. Furthermore, nanomechanical testing shows a large hardness of 14 GPa in the as-deposited samples, with a slight increase after thermal annealing and almost negligible irradiation hardening. Theoretical modeling based on ab initio methodologies combined with Monte Carlo techniques predicts the formation of Cr and V rich second phase particles and points at equal mobilities of point defects as the origin of the exceptional radiation tolerance. The fact that these alloys are suitable for bulk production coupled with the exceptional radiation and mechanical properties makes them ideal structural materials for applications requiring extreme conditions.

Sound event detection (SED) aims to detect when and recognize what sound events happen in an audio clip. Many supervised SED algorithms rely on strongly labelled data which contains the onset and offset annotations of sound events. However, many audio tagging datasets are weakly labelled, that is, only the presence of the sound events is known, without knowing their onset and offset annotations. In this paper, we propose a time-frequency (T-F) segmentation framework trained on weakly labelled data to tackle the sound event detection and separation problem. In training, a segmentation mapping is applied on a T-F representation, such as log mel spectrogram of an audio clip to obtain T-F segmentation masks of sound events. The T-F segmentation masks can be used for separating the sound events from the background scenes in the time-frequency domain. Then a classification mapping is applied on the T-F segmentation masks to estimate the presence probabilities of the sound events. We model the segmentation mapping using a convolutional neural network and the classification mapping using a global weighted rank pooling (GWRP). In SED, predicted onset and offset times can be obtained from the T-F segmentation masks. As a byproduct, separated waveforms of sound events can be obtained from the T-F segmentation masks. We remixed the DCASE 2018 Task 1 acoustic scene data with the DCASE 2018 Task 2 sound events data. When mixing under 0 dB, the proposed method achieved F1 scores of 0.534, 0.398 and 0.167 in audio tagging, frame-wise SED and event-wise SED, outperforming the fully connected deep neural network baseline of 0.331, 0.237 and 0.120, respectively. In T-F segmentation, we achieved an F1 score of 0.218, where previous methods were not able to do T-F segmentation., Comment: 12 pages, 8 figures

Hydrodynamics calculations have been successfully used in studies of the bulk properties of the Quark-Gluon Plasma, particularly of elliptic flow and shear viscosity. However, there are areas (for instance event-by-event simulations for flow fluctuations and higher-order flow harmonics studies) where further advancement is hampered by lack of efficient and precise 3+1D~program. This problem can be solved by using Graphics Processing Unit (GPU) computing, which offers unprecedented increase of the computing power compared to standard CPU simulations. In this work, we present an implementation of 3+1D ideal hydrodynamics simulations on the Graphics Processing Unit using Nvidia CUDA framework. MUSTA-FORCE (MUlti STAge, First ORder CEntral, with a~slope limiter and MUSCL reconstruction) and WENO (Weighted Essentially Non-Oscillating) schemes are employed in the simulations, delivering second (MUSTA-FORCE), fifth and seventh (WENO) order of accuracy. Third order Runge-Kutta scheme was used for integration in the time domain. Our implementation improves the performance by about 2~orders of magnitude compared to a~single threaded program. The algorithm tests of 1+1D~shock tube and 3+1D~simulations with ellipsoidal and Hubble-like expansion are presented., Comment: 3 pages, 3 figures, SQM2015 proceedings submitted to Journal of Physics: Conference Series (JPCS) - IOP Conference Series

We present our studies on jet-induced modifications of the characteristic of the bulk nuclear matter. To describe such a matter, we use efficient relativistic hydrodynamic simulations in (3+1) dimensions employing the Graphics Processing Unit (GPU) in the parallel programming framework. We use Cartesian coordinates in the calculations to ensure a high spatial resolution that is constant throughout the evolution of the system. We show our results on how jets modify the hydrodynamics fields and discuss the implications., Comment: 2 pages, 3 figures, SQM2015 proceedings submitted to Journal of Physics: Conference Series (JPCS) - IOP Conference Series

Understanding event-by-event correlations and fluctuations is crucial for the comprehension of the dynamics of heavy ion collisions. Relativistic hydrodynamics is an elegant tool for modeling these phenomena; however, such simulations are time-consuming, and conventional CPU calculations are not suitable for event-by-event calculations. This work presents a feasibility study of a new hydrodynamic code that employs graphics processing units together with a general MUSTA-FORCE algorithm (Multi-Stage Riemann Algorithm - First Order Centered scheme) to deliver a high-performance yet universal tool for event-by-event hydrodynamic simulations. We also investigate the performance of selected slope limiters that reduce the amount of numeric oscillations and diffusion in the presence of strong discontinuities and shock waves. The numerical results are compared to the exact solutions to assess the code's accuracy., Comment: 22 pages, 10 figures, preprint draft to International Journal of Nonlinear Sciences and Numerical Simulation

The understanding of Boolean automata networks dynamics takes an important place in various domains of computer science such as computability, complexity and discrete dynamical systems. In this paper, we make a step further in this understanding by focusing on their cycles, whose necessity in networks is known as the brick of their complexity. We present new results that provide a characterisation of the transient and asymptotic dynamics, i.e. of the computational abilities, of asynchronous Boolean automata networks composed of two cycles that intersect at one automaton, the so-called double-cycles. To do so, we introduce an efficient formalism inspired by algorithms to define long sequences of updates, that allows a better description of their dynamics than previous works in this area.

In order for counselors to more effectively serve their clients and contribute to an environment that is life enhancing for women and girls as well as for men and boys they must advocate an end to all violence against women. In a national study of women who escaped from battering, participants rated the effectiveness of formal help sources from very effective to somewhat effective. Results from the study reveal that participants selected women's groups and battered women's shelters as more effective than counseling or social services. These results indicate that counselors have much work to do if they are to effectively serve clients and help end violence within society. This paper discusses domestic violence and provides statistical information concerning its occurrence. Social advocacy strategies, client advocacy, and legislative advocacy in relation to domestic violence are discussed. Counselors are encouraged to act as advocates for victims of domestic violence and to continue to push for the eradication of violence against women. A list of recommended readings and a list of websites, organizations, and hotlines are provided. (Contains 10 references.) (MKA)