Parallel Way Research Articles

Skeleton-Based Action Recognition Through Contrasting Two-Stream Spatial-Temporal Networks

For pursuing accurate skeleton-based action recognition, most prior methods use the strategy of combining Graph Convolution Networks (GCNs) with attention-based methods in a serial way. However, they regard the human skeleton as a complete graph, resulting in less variations between different actions (e.g., the connection between the elbow and head in action “clapping hands”). For this, we propose a novel Contrastive GCN-Transformer Network (ConGT) which fuses the spatial and temporal modules in a parallel way. The ConGT involves two parallel streams: Spatial-Temporal Graph Convolution stream (STG) and Spatial-Temporal Transformer stream (STT). The STG is designed to obtain action representations maintaining the natural topology structure of the human skeleton. The STT is devised to acquire action representations containing the global relationships among joints. Since the action representations produced from these two streams contain different characteristics, and each of them knows little information of the other, we introduce the contrastive learning paradigm to guide their output representations of the same sample to be as close as possible in a self-supervised manner. Through the contrastive learning, they can learn information from each other to enrich the action features by maximizing the mutual information between the two types of action representations. To further improve action recognition accuracy, we introduce the Cyclical Focal Loss (CFL) which can focus on confident training samples in early training epochs, with an increasing focus on hard samples during the middle epochs. We conduct experiments on three benchmark datasets, which demonstrate that our model achieves state-of-the-art performance in action recognition.

Particle Swarm–Grey Wolf Cooperation Algorithm Based on Microservice Container Scheduling Problem

In recent years, microservices have been very widely used as a new application development technology in edge computing, IoT, and cloud computing. Application containerization technology is one of its core technologies, which allows multiple containers to be deployed within the same physical node. Then a single physical node could provide different services to user. How to rationally deploy containers on a cluster of physical nodes is one of the main research directions nowadays. Although a number of researchers have modeled the microservice container scheduling problem and proposed effective solutions, there are still shortcomings, for example, the slow speed of finding the optimal solution and the tendency of the algorithm to fall into local optimality. This paper propose a Particle Swarm - Grey Wolf Cooperation Algorithm based on Microservice Container Scheduling Problem (PS-GWCA) by using particle swarm optimization algorithm (PSO) and grey wolf algorithm (GWO) in a multi-core parallel way, which enables the two algorithms to complement each other in the whole search process through the information exchange between populations. In the early of the search stage, the GWO can use its global search capability to guide the PSO to jump out of the local optimum to avoid premature convergence, and in the late of the search stage, the PSO can enhance the search capability of the GWO on the pareto optimal frontier. The experimental results show that compared with the other three algorithms, the algorithm optimizes 18.07% in network transmission cost, 14.67% in local load balancing, 20.66% in global load balancing, and 7.5% in search speed, and 5.69% in service reliability.

A Novel Deep Parallel Time-Series Relation Network for Fault Diagnosis

Currently, deep learning based methods are widely used in the fault diagnosis of time series data for their high precision. However, the application of traditional deep learning fault diagnosis models is limited by their calculational efficiency and poor interpretation ability. To address the problems, a fault diagnosis model named the deep parallel time-series relation network (DPTRN) is proposed in this paper. There are three main advantages of the DPTRN. (1) Our proposed time relationship unit can perform feature extraction on each time node of a time series sample simultaneously, therefore, the DPTRN performs fault diagnosis in a parallel way and improves the computing efficiency significantly. (2) By improving the absolute position embedding, our novel decoupling position embedding unit can be directly applied for fault diagnosis and can learn contextual information. (3) Our proposed DPTRN has an obvious advantage in feature interpretability compared with traditional deep learning based models. Applying the DPTRN model on four datasets, we achieved higher diagnosis performance with much lower cost, which indicates the effectiveness, efficiency and interpretability of the proposed DPTRN model.

Monte Carlo solution of heat conductivity problems with quadratic nonlinearity on the boundary of domain

In this work, we study the approach connected with branching Markov processes. Branching Markov random processes are created to solve boundary-value problems (BVR) with polynomial nonlinearities. The realization of these processes creates so called “trees”. Unbiased estimators of the solution of the nonlinear problem are constructed on these random trees. We also calculated in parallel way the variance (statistical error) of the constructed unbiased estimators. We offer computational algorithms for solving some nonlinear diffusion problems which frequently appears in engineering problems, particularly in heat conductivity. The case of a quadratic nonlinearity for the boundary BVP is considered in detail. Algorithms, for diffusion BVPs with nonlinear boundary conditions, differ from proposed for linear diffusion BVPs algorithms early and we compare their efficiencies. The described algorithms applied to the computational problem of thermal engineering in the presence of a nonlinear boundary condition. Also, this problem is solved as a boundary value problem of conductive heat transfer.

Keys to Build an Inclusive University System: The Case of Spanish Public Universities

Sustainable Development Goals and the targets described in each of them can only be fully achieved with the commitment and involvement of universities. These institutions are the perfect environment to promote and enhance sustainable human development, in such a way that from their own structure and operation they favor the transition towards the building of peaceful, fair, and inclusive societies. To achieve the goals described in SDG4, universities must address three areas in a combined and parallel way: “Institutional Management and Governance”, “Education”, and “Leadership”. Following the development of four quantitative and qualitative studies that address these areas individually, the main common conclusions are presented, aimed at showing the path that the university system must follow to achieve these goals within 10 years. The results reveal the need to the focus on the development of an inclusive university culture based on the foundations of each institution (mission, vision, and values), as well as the need to strengthen the link with the NGO sector. Similarly, this inclusive culture must be transversal in university strategic plans to ensure that all actions developed in this context are aligned with the precepts of the 2030 Agenda.

MMS22L-TONSL functions in sister chromatid cohesion in a pathway parallel to DSCC1-RFC.

The leading strand-oriented alternative PCNA clamp loader DSCC1-RFC functions in DNA replication, repair, and sister chromatid cohesion (SCC), but how it facilitates these processes is incompletely understood. Here, we confirm that loss of human DSCC1 results in reduced fork speed, increased DNA damage, and defective SCC. Genome-wide CRISPR screens in DSCC1-KO cells reveal multiple synthetically lethal interactions, enriched for DNA replication and cell cycle regulation. We show that DSCC1-KO cells require POLE3 for survival. Co-depletion of DSCC1 and POLE3, which both interact with the catalytic polymerase ε subunit, additively impair DNA replication, suggesting that these factors contribute to leading-strand DNA replication in parallel ways. An additional hit is MMS22L, which in humans forms a heterodimer with TONSL. Synthetic lethality of DSCC1 and MMS22L-TONSL likely results from detrimental SCC loss. We show that MMS22L-TONSL, like DDX11, functions in a SCC establishment pathway parallel to DSCC1-RFC. Because both DSCC1-RFC and MMS22L facilitate ESCO2 recruitment to replication forks, we suggest that distinct ESCO2 recruitment pathways promote SCC establishment following either cohesin conversion or de novo cohesin loading.

Role of optical coherence tomography in diagnosis of Alzheimer's disease

AbstractPurpose: To evaluate changes in visual function and anatomical structures in the neuroretina of patients diagnosed with Alzheimer's disease (AD) in comparison with healthy controls.Methods: A total of 25 eyes of 13 AD patients and 80 eyes of 40 healthy subjects were analysed. Best corrected visual acuity (BCVA) was assessed in different contrast situations (100%, 2.5%, 1.25%) with ETDRS chart (Early Treatment Diabetic Retinopathy Study). Also, retinal nerve fibre layer (RNFL) and ganglion cell layer (GCL) thickness were analysed using Spectralis and Triton optical coherence tomography (OCT), globally and by sectors, both in peripapillary and macular area.Results: A statistically significant decrease was found in BCVA in AD patients in comparison with controls (p < 0.001), specifically in 100% and 1.25% contrast charts. Regarding structural analysis, a significant reduction in retinal thickness was found in AD patients in comparison with controls. Global thickness was found to be significantly thinner in peripapillary retina in Triton OCT (p < 0.001). In a similar way, macular GCL, both in Triton and Spectralis OCT, showed a significant thinning (p < 0.001) in global and sectorial analysis. RNFL showed a significant tendency towards thinning, but specifically in infero‐temporal sector of macular area in Spectralis OCT (p < 0.001).Conclusions: The neurodegeneration that AD produces to the brain affects the retina in a parallel way, and that can be objectified by measuring BCVA parameters and thickness measurements in OCT. Compared to healthy controls, AD has a significant thinning in several neuroretinal layers, which can be considered a potential diagnostic tool considering that OCT is a quick, reproducible, non‐invasive test.

Possibilities enhanced and constrained: Where dynamic semiosis works

The newly established field of possibility studies is set to solve complex theoretical problems created by the future-based setup of the notion of possibility. Irreversible time sets strict limits upon the uses of the concept of possibility and calls for its conceptualization in terms of cultural psychology of dynamic semiosis. Use of signs work in two parallel ways—organizing the present experience and catalyzing future possibilities. Negotiation of the constraint borders for the future is the main function of possibilities discourse. Hence investigation into possibilities requires coordination of intentionality of the actor with imagination about the future.

Parallel optical coherent dot-product architecture for large-scale matrix multiplication with compatibility for diverse phase shifters.

Photonics physically promises high-speed and low-consumption computing of matrix multiplication. Nevertheless, conventional approaches are challenging to achieve large throughput, high precision, low power consumption, and high density simultaneously in a single architecture, because the integration scale of conventional approaches is strongly limited by the insertion loss of cascaded optical phase shifters. Here, we present a parallel optical coherent dot-product (P-OCD) architecture, which deploys phase shifters in a fully parallel way. The insertion loss of phase shifters does not accumulate at large integration scale. The architecture decouples the integration scale and phase shifter insertion loss, making it possible to achieve superior throughput, precision, energy-efficiency, and compactness simultaneously in a single architecture. As the architecture is compatible with diverse integration technologies, high-performance computing can be realized with various off-the-shelf photonic phase shifters. Simulations show that compared with conventional architectures, the parallel architecture can achieve near 100× higher throughput and near 10× higher energy efficiency especially with lossy phase shifters. The parallel architecture is expected to perform its unique advantage in computing-intense applications including AI, communications, and autonomous driving.

Evaluating the synergistic effect of maple leaves extract and iodide ions on corrosion inhibition of Q235 steel in H2SO4 solution

In recent years, many plant extracts have been used as green corrosion inhibitors to replace traditional corrosion inhibitors, but their inhibition performance still needs to be improved. The synergistic corrosion inhibition behavior and mechanism of maple leaves extract (MLE) and potassium iodide (KI) on Q235 steel in 0.5 M H2SO4 solution were firstly investigated in this work. The results demonstrate that both MLE and KI can prevent the corrosion of steel in corrosive media, and that their combined protection effect is superior. The corrosion inhibition efficiency (η) value reaches 93.4 % in corrosive solution with the combination of 200 mg/L MLE and 200 mg/L KI, while the maximum of η is only 81.6 % using MLE alone. In addition, Fe-N bond and Fe-O bond were found on the steel surface through the X-ray photoelectron spectroscopy (XPS) analysis, which confirm the formation of a protective film. The adsorption behavior of an inhibitory film composed of MLE and KI mixtures onto steel surface follows the Langmuir adsorption isotherm model, consisting of physisorption and chemisorption. Finally, molecular dynamics (MD) simulation indicates that inhibitor molecules can adsorb on the Fe substrate in a parallel way with high binding energy values, which supports the experiments well.

Robust self-triggered DMPC for linear discrete-time systems with local and global constraints

In this paper, a novel robust distributed model predictive control with a self-triggered mechanism is proposed for a family of discrete-time linear systems with additive disturbances, local and global constraints. To handle the additive disturbances, a tube method is applied to achieve the robust satisfaction of local constraints. Meanwhile, a sequential constraints tightening method is proposed to guarantee the satisfaction of global coupled constraints. The optimization problem is constructed as a consensus problem of an augmented Lagrange function and is solved through a modified distributed alternative direction multiplier method. Furthermore, a self-triggered mechanism is adopted to help reduce the computation burden and accelerate system stabilization by skipping insignificant iteration steps in parallel ways. By clarifying the upper bound on tolerable bounded disturbances, two sufficient conditions about recursive feasibility of optimization problem and input-to-state stability of the closed-loop system are given, respectively. Finally, performances of the proposed scheme are illustrated by simulation examples.

Management of Intra-Uterine Growth Restriction by Ayurvedic Regime: A Case Report

Maternal nutrition and congenital malformation are considered as very important factors for intra uterine growth restriction. Present case deals with these two factors in parallel way. A 32 years young lady second gravida, G2P0A1, housewife, residence of Jaipur, came to NIA OPD with complaint of amenorrhea since a month. According to patient her last menstrual period was on 3/3/20. Her antenatal care was continued in NIA OPD. First fetal wellbeing scan reveals mild ventriculomegaly so she was advised for second level scan and quadruple marker but due to low socioeconomic condition patient was not able to do second level anomaly scan and she continued her pregnancy on her own risk. After 7 months she came to us with the complaints of decreased fetal movements. Fundal height corresponding to period of gestation and fetal heart sound was present which was regular. A color doppler was advised which shows normal fetoplacental and uteroplacental circulation, fetal weight was normal according to period of gestation. On the eighth month the fundal height was found decreased along with maternal weight decreased by 1kg. She was suspected for IUGR and planned for Ksheerbasti. After four episodes Ksheer basti on the 36 weeks of gestation, mother had significant weight gain of 2 kg along with increase in fundal height. USG was done which reveals Single live intra uterine pregnancy of 33 weeks 2 days with lumbar spina bifida with meningomyelocele, CHIARI-II brain malformation, EFW– 2282gm. She delivered a live female child of 2300gm through normal vaginal delivery at maternity hospital. Child was having severe form of myelomeningocele with brain malformation.

Coextension Paths and Access Paths: A Comparative Study of Abstract Motion in English and Arabic

This paper examines the elements of abstract motion as represented by coextension path expressions and access path expressions. The elements include Path, Manner, Duration, and Mover. It does so by comparing English and Arabic languages with the purpose of building a model for abstract motion in Arabic and of performing a comparative analysis of the linguistic representation of abstract motion in the two languages under consideration. The study found that the two types of abstract motion under consideration differ in how they express the four components of motion, namely, path, manner, duration, and mover. They also differ in the degree of specificity of each component and in perspective. Moreover, this paper attempted to draw a comparison between English and Arabic expression of abstract motion and found out that abstract motion represented by access and coextension path expressions exist in both languages in parallel ways except for some slight differences on the grammatical and semantic levels. It is discovered that abstract motion in both languages has a cognitive bias towards path and distance. This applies to both English and Arabic. There are differences between English and Arabic in the type of verbs used to realize abstract motion. Arabic uses a limited number of verbs in realizing abstract motion, which is different from verbs used for actual motion. Finally, the study is hoped to be useful in the characterization of abstract motion in Arabic and in understanding the difference between access paths and extension paths.

A Generic Battery Cycling Optimization Framework with Learned Sampling and Early Stopping Strategies

Battery optimization is a challenging task due to the large amount of cost and time required to evaluate different configurations in experiment or simulation. It is especially costly to optimize cycling performance since cycling the batteries is costly and time-consuming. In this project, we report an optimization framework with learned sampling and early stopping strategies, designed to optimize battery parameters in an efficient way to reduce total cycling time. It consists of a pruner and a sampler. The pruner stops unpromising cycling batteries to save the budget for further exploration. The sampler can predict the next promising configurations based on query history. The framework can deal with categorical, discrete and continuous parameters, and can run in an asynchronously parallel way to allow multiple simultaneous cycling cells. We demonstrated the performance by a parameter fitting problem for a calendar aging model. Comparisons with current methods are made to demonstrate the effectiveness of our method to reduce cycling time. Our framework can foster battery optimization in simulations and experiments.

Dynamics of DNA Origami Lattices.

Hierarchical assembly of programmable DNA frameworks─such as DNA origami─paves the way for versatile nanometer-precise parallel nanopatterning up to macroscopic scales. As of now, the rapid evolution of the DNA nanostructure design techniques and the accessibility of these methods provide a feasible platform for building highly ordered DNA-based assemblies for various purposes. So far, a plethora of different building blocks based on DNA tiles and DNA origami have been introduced, but the dynamics of the large-scale lattice assembly of such modules is still poorly understood. Here, we focus on the dynamics of two-dimensional surface-assisted DNA origami lattice assembly at mica and lipid substrates and the techniques for prospective three-dimensional assemblies, and finally, we summarize the potential applications of such systems.

Polarized discrete ordinate adding approximation for infrared and microwave radiative transfer

A POLarized Discrete ordinate aDding Approximation (POLDDA) is proposed to solve the vector thermal radiative transfer equation in a vertically inhomogeneous plane-parallel atmosphere. The discrete ordinates method is employed to find the single-layer solution for vector thermal radiative transfer; this is an extension of the approach for scalar radiative transfer, with the I- and Q-components being dealt with in a parallel way by using a doubled dimension in the discrete ordinate space. Chandrasekhar’s invariance principle is used to drive the adding process for the connection of multiple layers. This adding process has no restriction on the layer optical properties. The accuracy and efficiency of POLDDA are evaluated under different atmospheric conditions for different number of computational quadrature angles (streams). It is shown that the relative differences between POLDDA (using 32 or 16 streams) and the benchmark calculations (PolRadtran/RT3 with 32 streams, a typical model based on the adding-doubling method) are negligible for both I- and Q-components. When 8 streams are used, the accuracy of POLDDA slightly decreases, but still higher than that of 8-stream RT3. The computational efficiency of POLDDA is much higher than RT3, especially for large optical depths and when a large number of streams (≥ 16) are used. Unlike RT3, the computational time of POLDDA is always the same regardless of the optical depth.

Efficient Design of Thermoelastic Structures Using a Krylov Subspace Preconditioner and Parallel Sensitivity Computation

The repeated updating of parametric designs is computationally challenging, especially for large-scale multi-physics models. This work is focused on proposing an efficient modal modification method for gradient-based topology optimization of thermoelastic structures, which is essential when dealing with their complex eigenproblems and global sensitivity analysis for a huge number of design parameters. The degrees of freedom of the governing equation of thermoelastic structures is very huge when its parametric partial differential equation is discretized using the numerical technique. A Krylov subspace preconditioner is constructed based on the Neumann series expansion series so that the thermoelastic eigenproblem can be solved in an efficient low-dimension solver, rather than its original high-fidelity solver. In the construction of Krylov reduced-basis vectors, the computational cost of the systemic matrix inverse becomes a critical issue, which is solved efficiently by means of constructing a diagonal systemic matrix with the lumped mass and heat generation submatrices. Then, the reduced-basis preconditioner can provide an efficient optimal solver for both the thermoelastic eigenproblem and its eigen sensitivity. Furthermore, a master-slave pattern parallel method is developed to reduce the computational time of computing the global sensitivity numbers, and therefore, the global sensitivity problem can be efficiently discretized into element-scale problems in a parallel way. The sensitivity numbers can thus be solved at the element scale and aggregated to the global sensitivity number. Finally, two case studies of the iterative topology optimization process, in which the proposed modal modification method and the traditional method are implemented, are used to illustrate the effectiveness of the proposed method. Numerical examples show that the proposed method can reduce the computational cost remarkably with acceptable accuracy.

Large-Scale Subspace Clustering by Independent Distributed and Parallel Coding.

Subspace clustering is a popular method to discover underlying low-dimensional structures of high-dimensional multimedia data (e.g., images, videos, and texts). In this article, we consider a large-scale subspace clustering (LS2C) problem, that is, partitioning million data points with a millon dimensions. To address this, we explore an independent distributed and parallel framework by dividing big data/variable matrices and regularization by both columns and rows. Specifically, LS2C is independently decomposed into many subproblems by distributing those matrices into different machines by columns since the regularization of the code matrix is equal to a sum of that of its submatrices (e.g., square-of-Frobenius/ l1 -norm). Consensus optimization is designed to solve these subproblems in a parallel way for saving communication costs. Moreover, we provide theoretical guarantees that LS2C can recover consensus subspace representations of high-dimensional data points under broad conditions. Compared with the state-of-the-art LS2C methods, our approach achieves better clustering results in public datasets, including a million images and videos.

An Analysis on Statutory and Customary Land Ownership in Cameroon: Two Parallel Ways

Land is essential for every human activity on earth as it is the source of all material wealth. In order to regulate the ownership, use and development of land and land resources, nations over the world have instituted land ownership systems aimed at consistent balancing of the interests of the government, the land owning class and the landless class. This paper examines the parallel nature of statutory and customary land ownership in Cameroon, how it has evolved over the years until 1974 when a single land law, otherwise known as the 1974 Land Ordinance, which were established to harmonize and regulate land ownership in the country. The paper further contends that the present land ownership system in Cameroon as found in the ordinance is advocating inclinations with excessive state control of land ownership, use and development. In connection with our above objectives, we adopted an in-depth content analysis, which is based on primary and secondary sources of data collection imperative to the study. The paper concludes that such land system cannot effectively support the indigenes and development initiatives as it creates too much bureaucracy in land transactions, land registration and land titling. It recommends an urgent balance between the statutory and customary land ownership to facilitate access to land with ease for various purposes.

Radar Phase-Coded Waveform Design with Local Low Range Sidelobes Based on Particle Swarm-Assisted Projection Optimization

In modern electronic warfare, cognitive radar with knowledge-aided waveforms would show significant flexibility in anti-interference. In this paper, a novel method, named particle swarm-assisted projection optimization (PSAP), is introduced to design phase-coded waveforms with multi-level low range sidelobes, which mainly considers the stability for randomized initialization under the unimodular constraint. Firstly, the mathematical problem corresponding to avoid the range sidelobe masking from multiple non-cooperative targets or interference is formulated by giving different threat levels. Then, based on the alternating direction decomposition idea, the original problem is divided into triple-variable ones where these non-linear approximations can be solved via alternating projections along with FFT. Furthermore, the PSAP method with swarm intelligence, learning factor, and particle-assisted projection could ensure the optimization convergence in a parallel way, which could relax the non-convex constraint and enhance the global exploiting performance. Finally, simulations for several typical scenarios and numerical results are all provided to assess the waveforms generated by PSAP and other prevalent ones.