Search

We study an optimized Adaptive Monte Carlo algorithm for the Wigner kernel - an important problem in quantum mechanics. We will compare the results with the basic adaptive approach and other stochastic approaches for computing the Wigner kernel represented by difficult multidimensional integrals in dimension d up to 12. The higher cases d > 12 will be considered for the first time. A comprehensive study and an analysis of the computational complexity of the optimized Adaptive MC algorithm under consideration has also been presented.

An optimized Monte Carlo approach (OPTIMIZED MC) for a Fredholm integral equations of the second kind is presented and discussed in the present paper. Numerical examples and results are discussed and MC algorithms with various initial and transition probabilities are compared.

Recently stochastic methods have become very important tool for high performance computing of very high dimensional problems in computational finance. The advantages and disadvantages of the different highly efficient stochastic methods for multidimensional integrals related to evaluation of European style options will be analyzed. Multidimensional integrals up to 100 dimensions related to European options will be computed with highly efficient optimized lattice rules.

In this work we investigate advanced stochastic methods for solving a specific multidimensional problem related to neural networks. Monte Carlo and quasi-Monte Carlo techniques have been developed over many years in a range of different fields, but have only recently been applied to the problems in neural networks. As well as providing a consistent framework for statistical pattern recognition, the stochastic approach offers a number of practical advantages including a solution to the problem for higher dimensions. For the first time multidimensional integrals up to 100 dimensions related to this area will be discussed in our numerical study.

Organic electrochemical transistors (OECTs) have received significant attention especially in biomedical applications. Despite many efforts on modeling these transistors, simulating OECT-based circuits is still a challenge due to the absence of accurate mathematical models. In this paper a DC model for p-type depletion-mode OECTs is proposed that more closely mimics their characteristics compared to the Bernards–Malliaras (B–M) model. Although OECTs are mostly used as transconductance amplifiers, their use as voltage amplifiers is investigated here with measurements at various drain-source voltages. Compared to the B–M model, the proposed model has better matching up to 3.6% between simulations and measurements of the analyzed transistors.

Accurate and time-effective simulation and design optimization of quasi-optical (QO) systems is an extremely challenging electromagnetic problem given the multi-scale dimension of QO components and the need to consider the finite size of such components to account for effects like diffraction. To show that the Method of Moments (MoM) provides an elegant solution for these problems, truncated rectangular arrays of tilted slots are measured in a QO Terahertz (THz) time-domain setup and comparison with MoM is carried out. The extraordinary transmission peaks are modulated by the size of the array and the orientation of the slots with respect to the incident electric field.

Stochastic techniques have been developed over many years in a range of different fields, but have only recently been applied to the problems in machine learning. A fundamental problem in this area is the accurate evaluation of multidimensional integrals. An introduction to the theory of the stochastic optimal generating vectors has been given. A new optimized lattice sequence with a special choice of the optimal generating vector has been applied to compute multidimensional integrals up to 30-dimensions. Clearly, the progress in the area of machine learning is closely related to the progress in reliable algorithms for multidimensional integration.

In this paper we study numerically an optimized Adaptive Monte Carlo algorithm for the Wigner kernel - an important problem in quantum mechanics represented by difficult multidimensional integrals. We will show the advantages of the optimized Adaptive MC algorithm and compare the results with the Adaptive approach from our previous work [4] and other stochastic approaches for computing the Wigner kernel in 3,6,9dimensional case. The 12-dimensional case will be considered for the first time. A comprehensive study and an analysis of the computational complexity of the optimized Adaptive MC algorithm under consideration has also been presented.

The results from an analysis of the influence of force majeure on the continuity of electricity supply in medium voltage electricity distribution networks in the European Union and Bulgaria are presented in the paper. For this purpose, the influence of force majeure circumstances on the quality indicators of the electric power supply SAIDI and SAIFI for two regions in Bulgaria - a region with a predominant cable network and a region with a predominant overhead network - is considered.

The results of a study of the levels of current and voltage harmonics, in a power supply system feeding a vertical mill for cement grinding, are presented in this paper. The obtained results are analyzed and compared with regulations.

Results for the indicators of power supply continuity in the electrical distribution networks (EDN) of two typical regions in the Republic of Bulgaria. The purpose of the study is to analyze and evaluate the impact of the type of electrical distribution networks on the indicators of quality of electricity supply in medium voltage distribution networks and to determine the possibilities for reducing the indicators of power supply interruptions.

Since the appearance of Data Mining concept until the era of Big Data, the change in size of analyzed data, the arise of high performance systems, new technologies and software programs. Data science emerge as a field, which combines many approaches and techniques, related to data analyzing, process of data cleaning, integration of data, selection and transformation, knowledge extraction, evaluation and representation. The report outlines the main problems the authors encountered in cleaning up the data before analyzing it and proposes new SCSV software (scala comma separated values). Experiments have been conducted to clean up and analyze large volumes of data, based on machine learning techniques, from annual surveys among StackOverflow users.

The results from analysis of the indicators of continuity of power supply in electricity distribution grids medium voltage are presented in the paper. Measures for improving of the indicators under consideration are proposed.

Online controlled experiments are the gold standard for evaluating improvements and accelerating innovations in online and app worlds. However, little is known about applicability, implementation, and efficacy of experimentation for operating systems (OS), where many features are non-user-facing. In this paper, we present the Windows Experimentation platform (WExp), and insights from implementation and execution of real-world experiments in the OS. We start by discussing the need for experimentation in OS, using real experiments to illustrate the benefits. We then describe the architecture of WExp, focusing on unique considerations in its engineering. Finally, we discuss learnings and challenges from conducting real-world experiments. Our experiences and insights can motivate practitioners to start experimenting as well as to help them to successfully build their experimentation platforms. The learnings can also guide experimenters with best-practices and highlight promising avenues for future research.

There is a continuing need for higher density routing and heterogeneous integration of different devices, but also for continuous cost reduction. While traditional organic flip-chip substrates using semi-additive processes (SAP) have not been able to scale to ultra-fine DL pitches and via opening below 10 $\mu$m, photo-sensitive spin-on dielectrics and RDL processes used for wafer level packaging do not sufficiently address the cost reduction need, and also face serious technical challenges. Therefore, the integrity and reliability of the pattern becomes necessary as feature sizes shrink to below 5 $\mu$m, demanding innovative thinking on the process integration to overcome these barriers. An innovative fabrication process has been developed for RDL and via formation using excimer laser ablation of an optimized dielectric material for each layer. Metallization is a key factor in making this process practical and cost effective. Yield and reliability is dependent on the adhesion between seed layer and dielectric film. Precise and optimized bottom-up plating is required to eliminate the need for a chemical mechanical planarization (CMP) process. This paper presents a novel excimer laser enabled dual damascene process for ultra-fine routing for BEOL. Fine RDL down to 2/2$\mu$m L/S and vias down to 5$\mu$m diameter are patterned using an excimer laser stepper with a reticle mask. We will report progress in demonstrating fabrication and technical characterization of this new embedded trace approach with and without planarization by CMP or fly-cutting methods. Capabilities and limitations of this bottom-up plating process will be discussed, its technical robustness and commercial advantages are demonstrated, and topics for further optimization are highlighted.

A new Monte Carlo algorithm for solving systems of Linear Algebraic (LA) equations is presented and studied. The algorithm is based on the “Walk on Equations” Monte Carlo method recently developed by Ivan Dimov, Sylvain Maire and Jean Michel Sellier [4]. The algorithm is optimized by choosing the appropriate values for the relaxation parameters which leads to dramatic reduction in time and lower relative errors for a given number of iterations. Numerical tests are performed for examples with matrices of different size and on a system coming from a finite element approximation of a problem describing a beam structure in constructive mechanics.

Business continuity and availability are top priority for IT. Providing sustainable availability, confidentiality and integrity of all services and resources within associates and counterparts is a key concept to be in the “Leaders” quadrant. Conventionally system updates are being underestimated, delayed or forcefully applied lacking any prior research, analysis nor testing. This paper would glance over an approach providing a flexible and consistent update deployment infrastructure and up-to-date clients, while assuring close to 0 business disruption and having latest critical and security updates in place protecting the organization from current exploits.

The continued adoption of fan-out structures for advanced packaging and the rise of heterogeneous integration of different devices require several layers of interconnects, with feature sizes shrinking below 5 microns. Traditional organic flip-chip substrates using the semi-additive process (SAP) have not been able to scale to ultra-fine RDL pitches and via openings below 10 microns. Photo-sensitive spin-on dielectrics and RDL processes used for wafer level packaging face serious technical challenges, causing reliability and pattern integrity concerns. Furthermore, both processes do not sufficiently address the need for cost reduction. An innovative dual damascene fabrication process has been developed to create embedded fan-out structures using excimer laser ablation. An optimized dielectric material for RDL layers is used that is capable of resolving features down to 2/2 microns line/space through a mask reticle with high throughput. Metallization is a key factor in making this process cost effective. Yield and reliability are dependent on excellent adhesion between seed layer and dielectric film. Precise and optimized bottom-up plating as well as good within-wafer uniformity is required to eliminate the need for a chemical mechanical planarization (CMP) process. In this paper, we will report progress in demonstrating fabrication of multi-layer RDL with 2-5 micron trenches and vias enabled by the new embedded trace approach without requiring planarization by CMP or fly-cutting methods. We will present electrical properties together with mechanical and thermal reliability data of via and RDL traces patterned by this excimer laser enabled dual damascene process. The cost advantage over traditional fabrication processes will be highlighted.