Search

This paper presents a comprehensive study and analysis of the intelligent grouping of papers in adult higher education using a random matrix approach. Using the results of random matrix theory on the eigenvalues of the sample covariance matrix, the energy of each subspace is estimated, and the estimated energy is then used to construct a subspace weighting matrix. The statistical properties of the sample covariance matrix eigenvectors are analyzed using the first-order perturbation approximation, and then, asymptotic results from random matrix theory on the projection of the sample covariance matrix signal subspace to the real signal parametrization are used to obtain the weighting matrix based on the random matrix eigenvectors. Dynamic adjustment according to the fitness of individuals in the population is performed to ensure population diversity, while the combination of the small habitat technique can avoid the algorithm from falling into early convergence. The algorithm introduces chaos theory to optimize the population initialization process and uses the dynamic traversal randomness of chaos to select individuals in the population so that the initial population is close to the desired target solution. The design of the fitness function in the genetic algorithm generally maps the objective function of the problem to the fitness function. A good fitness function can directly reflect the quality of the individuals in the group. Based on the in-depth study of the basic attributes of the test questions and the principles of test paper evaluation, the mathematical model and objective function of intelligent paper grouping are determined by the difficulty, knowledge points, and cognitive level of the test questions as the main constraints, and NCAGA is applied to the intelligent paper grouping method, which better completes the intelligent paper grouping session for the computer system architecture course. In the process of designing the intelligent grouping algorithm, for the situations of premature convergence and convergence to locally optimal solutions that easily occur in the traditional genetic algorithm, this paper adopts the approach of adaptive adjustment of crossover probability and variation probability to improve the algorithm and achieves satisfactory results. Based on extensive business research, this paper completes the requirement analysis of the online practice system based on the intelligent grouping of papers and presents the functional design and database design of the key functional modules in the system in detail. Finally, this paper conducts functional tests on the system and analyses the test results. [ABSTRACT FROM AUTHOR]

Human social relationships imply conforming to the norms, behaviors, and cultural values of the society, but also socialization of emotions, to learn how to interpret and show them. In multiagent systems, much progress has been made in the analysis and interpretation of both emotions and norms. Nonetheless, the relationship between emotions and norms has hardly been considered and most normative agents do not consider emotions, or vice-versa. In this article, we provide an overview of relevant aspects within the area of normative agents and emotional agents. First we focus on the concept of norm, the different types of norms, its life cycle and a review of multiagent normative systems. Second, we present the most relevant theories of emotions, the life cycle of an agent’s emotions, and how emotions have been included through computational models in multiagent systems. Next, we present an analysis of proposals that integrate emotions and norms in multiagent systems. From this analysis, four relationships are detected between norms and emotions, which we analyze in detail and discuss how these relationships have been tackled in the reviewed proposals. Finally, we present a proposal for an abstract architecture of a Normative Emotional Agent that covers these four norm-emotion relationships. [ABSTRACT FROM AUTHOR]

Background and Objectives: Today, the increased number of Internet-connected smart devices require powerful computer processing servers such as cloud and fog and necessitate fulfilling requests and services more than ever before. The geographical distance of IoT devices to fog and cloud servers have turned issues such as delay and energy consumption into major challenges. However, fog computing technology has emerged as a promising technology in this field. Methods: In this paper, service/energy management approaches are generally surveyed. Then, we explain our motivation for the systematic literature review procedure (SLR) and how to select the related works. Results: This paper introduces four domains of service management and energy management, including Architecture, Resource Management, Scheduling management, and Service Management. Scheduling management has been used in 38% of the papers. Therefore, they have the highest service management and energy management. Also, Resource Management is the second domain that has been able to attract about 26% of the papers in service management and energy management. Conclusion: About 81% of the fog computing papers simulated their approaches, and the others implemented their schemes using a testbed in the real environment. Furthermore, 30% of the papers presented an architecture or framework for their research, along with their research. In this systematic literature review, papers have been extracted from five valid databases, including IEEE Xplore, Wiley, Science Direct (Elsevier), Springer Link, and Taylor & Francis, from 2013 to 2022. We obtained 1596 papers related to the discussed subject. We filtered them and achieved 47 distinct studies. In the following, we analyze and discuss these studies; then we review the parameters of service quality in the papers, and ultimately, we present the benefits, drawbacks, and innovations of each study. [ABSTRACT FROM AUTHOR]

This report describes The University of Texas Student Cluster Competition team’s effort to reproduce the results of “MemXCT: memory-centric X-ray CT reconstruction with massive parallelization” (Hidayetoğlu et al., 2019). The article details a new memory-centric approach that reconstructs X-ray computed tomography (XCT) from noisy raw data. In our reproduction experiments, we utilized Microsoft Azure’s CycleCloud tool to provision, orchestrate, and manage our computing cluster in the cloud. In particular, we scheduled and benchmarked reconstruction workloads using Azure’s CPU-based HC44rs and GPU-based NC12s v2 virtual machine (VM) types to evaluate the scalability properties of the reconstruction approach and the performance differences between architectures. The HC44rs VMs contained 44 Intel Xeon Platinum cores, while the NC12s v2 VM was equipped with two NVIDIA P100 GPUs. We used a recent version of Intel’s compiler stack with the MKL library for our CPU code along with CUDA 11.1 on GPUs. Overall, our results confirm the findings of the original article, demonstrating similar acceleration on GPUs and scalability properties on CPUs. Digital artifacts from these experiments are available at: 10.5281/zenodo.5598108 [ABSTRACT FROM AUTHOR]

A. Kireeva et al. contributed another paper in the field of cellular automata: "Parallel simulation of drift-diffusion-recombination by cellular automata and global random walk algorithm". Parallel computing technologies enable the solution of large-scale numerical simulation and data processing problems in science and industry. The previously proposed contention-aware scheduling algorithm is formally described with this model, modifications of the algorithm are proposed: look-ahead technique, chunked data transfers, data caching, and peer-to-peer data transfers. [Extracted from the article]

Parallel computing has become a crucial topic in the concern of computer science and also it is revealed to be critical when researching in high performance. The evolution of computer architectures towards an improved number of nodes, where parallelism could be the approach to option for speeding up an algorithm within the last few decades. The combination of processing units build a model of computation (circuits) has gained an essential place in the area of high performance computing (HPC) due to its configuration and considerable processing supremacy that is parallel, series, etc.... In this paper, we study the idea of parallel computing, and its programming models and also explore some theoretical and technical concepts which can be often needed to understand the Interconnection network. In particular, we show how this technology is new in assisting the field of computational physics, especially when the issue is data parallel. In this paper firstly we convert the graphical model of perfect difference network into circuit diagram then convert circuit diagram into switching function, then simplify it and redraw the equivalent network. [ABSTRACT FROM AUTHOR]

Network slicing is the key to enable virtualized resource sharing among vertical industries in the era of 5G communication. Efficient resource allocation is of vital importance to realize network slicing in real-world business scenarios. To deal with the high algorithm complexity, privacy leakage, and unrealistic offline setting of current network slicing algorithms, in this paper we propose a fully decentralized and low-complexity online algorithm, DPoS, for multi-resource slicing. We first formulate the problem as a global social welfare maximization problem. Next, we design the online algorithm DPoS based on the primal-dual approach and posted price mechanism. In DPoS, each tenant is incentivized to make its own decision based on its true preferences without disclosing any private information to the mobile virtual network operator and other tenants. We provide a rigorous theoretical analysis to show that DPoS has the optimal competitive ratio when the cost function of each resource is linear. Extensive simulation experiments are conducted to evaluate the performance of DPoS. The results show that DPoS can not only achieve close-to-offline-optimal performance, but also have low algorithmic overheads. [ABSTRACT FROM AUTHOR]

The article introduces a new section structure for the SIAM Journal on Scientific Computing (SISC), including a focus on Scientific Machine Learning as a major development in the field of scientific computing. Topics include the development of accurate, efficient, and scalable machine learning algorithms for solving mathematical modeling problems, emphasizing the importance of reproducibility in the review process.

The problem of resource-saving scheduling in a fog environment is considered in this paper. The objective function of the problem in question presupposes the fog nodes' reliability function maximizing. Therefore, to create a schedule, the following is required: the history of the fog devices' state changes and the search space, which consists of preselected nodes of the cloud-fog broker neighbourhood. The obvious approach to providing the scheduler with this information is to poll the fog nodes, yet this can consume the unacceptable time because of the QoS requirements. In this paper, the system architecture and general methods for efficient resource-saving scheduling is presented. The system is based on distributed ledger element usage, which provides the nodes with the proper awareness about the surroundings. The usage of the distributed ledger allows not only for the creation of the resource-saving schedule but also the reduction of the scheduling problem-solving time, which frees addition time that can be used for the solving of user tasks. The latter also affects the overall resource-saving via reliability. The novelty of this paper consists in the development of the hybrid ledger-based system, which integrates and arranges the elements of various ledger types to solve the newly formulated problem. [ABSTRACT FROM AUTHOR]

The current article discusses Sierpinski networks, which are fractal networks with certain applications in computer science, physics, and chemistry. These networks are typically used in complicated frameworks, fractals, and recursive assemblages. The results derived in this study are in mathematical and graphical format for particular classes of these networks of two distinct sorts with two invariants, K-Banhatti Sombor (KBSO) and Dharwad, along with their reduced forms. These results can facilitate the formation, scalability, and introduction of novel interconnection network topologies, chemical compounds, and VLSI processor circuits. The mathematical expressions employed in this research offer modeling insights and design guidelines to computer engineers. The derived simulation results demonstrate the optimal ranges for a certain network. The optimization assistant tool deployed in this work provides a single maximized value representing the maximum optimized network. These ranges can be put into service to dynamically establish a network according to the requirements of this paper. [ABSTRACT FROM AUTHOR]

This paper aims to propose innovative actions of advanced technological solutions and consequent necessary institutional decisions to achieve in a reasonable time the definitive confidential data protection and data sovereignty, based on available scientific results. Confidential data protection is a fundamental and strategic issue in next-generation Internet systems to guarantee data sovereignty and the respect of human rights as stated in the foundation of the United Nations. Even if presently many international regulations are decisive steps to guarantee data protection within normative contexts, they are not adequate to face new technologies, such as facial recognition, automatic profiling, position tracking, biometric data, AI applications, and many others in the future, as they are implemented without any awareness by the interested subjects. Therefore, a new approach to data protection is mandatory based on innovative and disruptive technological solutions. A recent OECD report highlighted the need for the so-called Privacy-Enhancing Technologies (PETs) for the effective protection of confidential data, even more urgent for the coexistence of privacy and data sharing in international contexts. A common feature of these technologies is the use of software methodologies that can run on currently available microprocessors and their present immaturity. More effective and definitive protection can be achieved with another methodological approach based on the paradigm of 'Data Usage Control'. This new concept guarantees data protection policy by default and initial design and it requires a new architecture of the data and a new HW&SW architecture of the computers. This contribution has a two-fold objective: first, to clarify why regulations alone and present technological proposals are not adequate for the effective and definitive protection of data and, second, to indicate the new necessary technological approach and the simultaneous institutional actions required to achieve the definitive protection and sovereignty of data in reasonable times, based on the results already available in the scientific literature. [ABSTRACT FROM AUTHOR]

In this paper, we first analyze a real storage system consisting of 72 SSDs utilizing either Hardware RAID (HW-RAID) or Software RAID (SW-RAID), and show that SW-RAID is up to 7× faster. We then reveal that with an increasing number of SSDs, the limited I/O parallelism in SAS controllers and multi-enclosure handshaking overheads cause a significant performance drop, minimizing the total I/O Per Second (IOPS) of a 144-SSD system to less than a single SSD. Second, we disclose the most important architectural parameters that affect a large-scale storage system. Third, we propose a framework that models a large-scale storage system and estimates the system IOPS and system resource usage for various architectures. We verify our framework against a real system and show its high accuracy. Lastly, we analyze a use case of a 240-SSD system and reveal how our framework guides architects in storage system scaling. [ABSTRACT FROM AUTHOR]

GPUs are widely used for diverse applications, particularly data-parallel tasks like machine learning and scientific computing. However, their efficiency is hindered by architectural limitations, inherited from historical RISC processors, in handling memory loads causing high register file contention. We observe that a significant number (around 26%) of values present in the register file are typically used only once, contributing to more than 25% of the total register file bank conflicts, on average. This paper addresses the challenge of single-use memory values in the GPU register file (i.e. data values used only once) which wastes space and increases latency. To this end, we introduce a novel mechanism inspired by CISC architectures. It replaces single-use loads with direct memory operands in arithmetic operations. Our approach improves performance by 20% and reduces energy consumption by 18%, on average, with negligible (<1%) hardware overhead. [ABSTRACT FROM AUTHOR]

Self-supervised learning has shown value for uncovering informative movement features for human activity recognition. However, there has been minimal exploration of this approach for affect recognition where availability of large labelled datasets is particularly limited. In this paper, we propose a P-STEMR (Parallel Space-Time Encoding Movement Representation) architecture with the aim of addressing this gap and specifically leveraging the higher availability of human activity recognition datasets for pain-level classification. We evaluated and analyzed the architecture using three different datasets across four sets of experiments. We found statistically significant increase in average F1 score to 0.84 for pain level classification with two classes based on the architecture compared with the use of hand-crafted features. This suggests that it is capable of learning movement representations and transferring these from activity recognition based on data captured in lab settings to classification of pain levels with messier real-world data. We further found that the efficacy of transfer between datasets can be undermined by dissimilarities in population groups due to impairments that affect movement behaviour and in motion primitives (e.g. rotation versus flexion). Future work should investigate how the effect of these differences could be minimized so that data from healthy people can be more valuable for transfer learning. [ABSTRACT FROM AUTHOR]

In today's digital landscape, organizations face significant challenges, including sensitive data leaks and the proliferation of hate speech, both of which can lead to severe consequences such as financial losses, reputational damage, and psychological impacts on employees. This work considers a comprehensive solution using a microservices architecture to monitor computer usage within organizations effectively. The approach incorporates spyware techniques to capture data from employee computers and a web application for alert management. The system detects data leaks, suspicious behaviors, and hate speech through efficient data capture and predictive modeling. Therefore, this paper presents a comparative performance analysis between Spring Boot and Quarkus, focusing on objective metrics and quantitative statistics. By utilizing recognized tools and benchmarks in the computer science community, the study provides an in-depth understanding of the performance differences between these two platforms. The implementation of Quarkus over Spring Boot demonstrated substantial improvements: memory usage was reduced by up to 80% and CPU usage by 95%, and system uptime decreased by 119%. This solution offers a robust framework for enhancing organizational security and mitigating potential threats through proactive monitoring and predictive analysis while also guiding developers and software architects in making informed technological choices. [ABSTRACT FROM AUTHOR]

Distributed public ledgers, the key to modern cryptocurrencies and the heart of many novel applications, have scalability problems. Ledgers such as the blockchain underlying Bitcoin can process fewer than 10 transactions per second (TPS). The cost of transactions is high, and the time to confirm a transaction is in the minutes. We present the BlockGraph, a scalable distributed public ledger inspired by principles of computer architecture. The BlockGraph exploits the natural locality of transactions to allow publishing independent transactions in parallel. It extends the blockchain with three new transactions to create a unified consistent ledger out of essentially independent blockchains. The most important change is the introduction of the blockstamp transaction, which essentially checkpoints a local blockchain and secures it against attack. The result is a locality-based, simple, secure, sharding protocol which keeps all transactions readable. This paper introduces the BlockGraph protocol, proves that it is consistent and can achieve many thousands of TPS. Using our implementation (a small extension to Bitcoin core) we demonstrate that it, in practice, can significantly improve throughput. [ABSTRACT FROM AUTHOR]

In this study, we obtain a sex prediction algorithm based on CNN in two ways - building a red Convolutional Neural Network (CNN) model from scratch and via transfer learning. We built a model from scratch and compared it with fine-tuned EfficientNetB1. We use these models for gender determination using periocular images and compare the two models depending on the accuracy of the models. The CNN model proposed from scratch yields an accuracy of 94.46% while the fine-tuned EfficientNetB1 yields an accuracy of 97.94%. This paper is one of the first works in determining gender from periocular images in the visible spectrum using a CNN model built from the outset. [ABSTRACT FROM AUTHOR]

This paper introduces an event‐based computing paradigm, where workers only perform computation in response to external stimuli (events). This approach is best employed on hardware with many thousands of smaller compute cores with a fast, low‐latency interconnect, as opposed to traditional computers with fewer and faster cores. Event‐based computing is timely because it provides an alternative to traditional big computing, which suffers from immense infrastructural and power costs. This paper presents four case study applications, where an event‐based computing approach finds solutions to orders of magnitude more quickly than the equivalent traditional big compute approach, including problems in computational chemistry and condensed matter physics. [ABSTRACT FROM AUTHOR]

The COVID-19 pandemic has brought about a profound transformation in the educational landscape in recent months. Educators worldwide have been challenged to tackle academic issues they could never have imagined. Among the most stressful situations faced by students and teachers is implementing online assessments. This paper proposes a system that includes exam prototypes for computer architecture modules at the higher education level. This system generates a wide range of questions and variations on the server side, supported by a set of simulators, resulting in many unique examination proposals. This system streamlines the monitoring process for the teacher, as it eliminates the possibility of two students receiving similar exams and reduces student stress by allowing them to practice with a limitless number of exam samples. This paper also highlights several indicators that demonstrate the advantages of this framework. [ABSTRACT FROM AUTHOR]

Background and Objectives: Big data is a combination of structured, semi-structured and unstructured data collected by organizations that must be stored and used for decision-making. Businesses that deal with the business intelligence system, as well as their data sources, have a major challenge in exploiting Big Data. The current architecture of business intelligence systems is not capable of incorporating and exploiting Big Data. In this paper, an architecture is developed to respond to this challenge. Methods: This paper focuses on the promotion of business intelligence to create an ability to exploit Big Data in business intelligence. In this regard, a new architecture is proposed to integrate both Business Intelligence and Big Data architectures. To evaluate the proposed architecture, we investigated business intelligence architecture and Big Data architecture. Then, we developed a Unified Modeling Language diagram for the proposed architecture. In addition, using the Colored Petri-Net, the proposed architecture is evaluated in a case study. Results: The results show that our architectural system has a higher efficiency in performing all steps, average time, and maximum time compared to business intelligence architecture. Conclusion: The proposed architecture can help companies and organizations gain more value from their data sources and better support managers and organizations in their decision-making. [ABSTRACT FROM AUTHOR]

The ancient Vedic mathematics is well known for quicker handy multiplications but its recognition as an integrated circuit core against existing hardware multipliers is not established. As optimized hardware implementation of binary multiplier is one of the prominent unsolved problems in computer architecture, this paper proposes efficient Urdhava Tiryakbhyam Vedic multiplier architecture and compares it with the set of hierarchical multiplication algorithms which generate multiplication result in a single clock cycle. Two innovative algorithms are proposed here, one with a compact structure and another for faster execution. Also, its optimized transistor level layout is designed and implemented. To maintain homogeneity for comparison, all the algorithms are programmed on a common HDL language platform and analyzed with the same tool and technology. Final results indicate that the proposed architecture delivers 15.5% less power delay product (PDP) compared to closest competitor algorithm. [ABSTRACT FROM AUTHOR]

The Internet of Drones (IoD) has become a hot research topic in academia, industry, and management in current years due to its wide potential applications, such as aerial photography, civilian, and military. This paper presents a comprehensive survey of IoD and its applications, deployments, and integration. We focused in this review on two main sides; IoD Applications include smart cities surveillance, cloud and fog frameworks, unmanned aerial vehicles, wireless sensor networks, networks, mobile computing, and business paradigms; integration of IoD includes privacy protection, security authentication, neural network, blockchain, and optimization based-method. A discussion highlights the hot research topics and problems to help researchers interested in this area in their future works. The keywords that have been used in this paper are Internet of Drones. [ABSTRACT FROM AUTHOR]

The ever-growing deluge of astronomical data challenges traditional server-based processing, hindering real-time analysis and scientific discovery. This paper proposes a novel approach: edge computing directly on an sCMOS camera using a System-on-Chip (SoC) architecture currently developed at Creotech Instruments. We present a custom-designed camera equipped with an FPGA-based SoC, enabling on-board preprocessing and feature extraction of astronomical images. This significantly reduces data transmission, minimizes latency, and empowers real-time decision-making for critical observations. We showcase the camera's capabilities through real-world scenarios, demonstrating its usability in astronomy. [ABSTRACT FROM AUTHOR]

The Peripheral Component Interconnect Express (PCI-Express or PCle) architecture has been the ubiquitous backbone interconnect in the evolving computing landscape for more than two decades. This paper delves into the multiple innovations driving the backward-compatible evolution of PCle for seven generations, doubling bandwidth every generation, while delivering power-efficient and cost-effective performance. Compute Express Link (CXL) overlays coherency and memory protocols on top of PCle for heterogeneous computing, addressing the memory wall, resource pooling and sharing across servers, and distributed computing using load-store based messaging. The die-todie industry-standard, Universal Chiplet Interconnect Express $^{TM}$ (UCle), offers orders of magnitude improvement in bandwidth density, power efficiency, and latency for PCle and CXL protocols on-package and pod-level connectivity with co-packaged optics. We foresee PCle will continue to evolve over the next few decades to serve future computing needs. It will do so by embracing alternate media while backward-compatible frequency scaling with copper will continue and protocol enhancements will enable non-tree fabric topologies. [ABSTRACT FROM AUTHOR]

Recent advances in smart connected vehicles and Intelligent Transportation Systems (ITS) are based upon the capture and processing of large amounts of sensor data. Modern vehicles contain many internal sensors to monitor a wide range of mechanical and electrical systems and the move to semi-autonomous vehicles adds outward looking sensors such as cameras, lidar, and radar. ITS is starting to connect existing sensors such as road cameras, traffic density sensors, traffic speed sensors, emergency vehicle, and public transport transponders. This disparate range of data is then processed to produce a fused situation awareness of the road network and used to provide real-time management, with much of the decision making automated. Road networks have quiet periods followed by peak traffic periods and cloud computing can provide a good solution for dealing with peaks by providing offloading of processing and scaling-up as required, but in some situations latency to traditional cloud data centres is too high or bandwidth is too constrained. Cloud computing at the edge of the network, close to the vehicle and ITS sensor, can provide a solution for latency and bandwidth constraints but the high mobility of vehicles and heterogeneity of infrastructure still needs to be addressed. This paper surveys the literature for cloud computing use with ITS and connected vehicles and provides taxonomies for that plus their use cases. We finish by identifying where further research is needed in order to enable vehicles and ITS to use edge cloud computing in a fully managed and automated way. We surveyed 496 papers covering a seven-year timespan with the first paper appearing in 2013 and ending at the conclusion of 2019. [ABSTRACT FROM AUTHOR]

In this paper, we consider the complicated dynamics of a delay-based photonic reservoir computing system. Since conventional computer architectures are approaching their limit, it is imperative to find new, efficient and fast ways of data processing. Photonic reservoir computing (RC) is a promising way which combines the computational capabilities of recurrent neural networks with high processing speed and energy efficiency of photonics. As the RC system is very promising, we analyze its dynamics so that we can make better use of it. In this paper, we mainly focus on its double Hopf bifurcation. We first analyze the existence of double Hopf bifurcation points. Then we use DDE-BIFTOOL to draw the bifurcation diagrams with respect to two bifurcation parameters, i.e. feedback strength η and delay τ , and give a clear picture of the double Hopf bifurcation points of the system. These figures show stability switches and the existence of double Hopf bifurcation points. Finally, we employ the method of multiple scales to obtain their normal forms, use the method of normal form to unfold and classify their local dynamics. The classification and unfolding of these double Hopf bifurcation points are obtained. Three types of double Hopf bifurcations are found. We verify the results by numerical simulations and find its complicated behavioral dynamics. For example, there exist stable equilibrium, stable periodic and quasi-periodic solutions in distinct regions. The discovered rich dynamical phenomena can help us to choose suitable values of parameters to achieve excellent performance of RC. [ABSTRACT FROM AUTHOR]

Computer vision (CV) algorithms have been extensively used for a myriad of applications nowadays. As the multimedia data are generally well-formatted and regular, it is beneficial to leverage the massive parallel processing power of the underlying platform to improve the performances of CV algorithms. Single Instruction Multiple Data (SIMD) instructions, capable of conducting the same operation on multiple data items in a single instruction, are extensively employed to improve the efficiency of CV algorithms. In this paper, we evaluate the power and effectiveness of RISC-V vector extension (RV-V) on typical CV algorithms, such as Gray Scale, Mean Filter, and Edge Detection. By our examinations, we show that compared with the baseline OpenCV implementation using scalar instructions, the equivalent implementations using the RV-V (version 0.8) can reduce the instruction count of the same CV algorithm up to 24x, when processing the same input images. Whereas, the actual performances improvement measured by the cycle counts is highly related with the specific implementation of the underlying RV-V co-processor. In our evaluation, by using the vector co-processor (with eight execution lanes) of Xuantie C906, vector-version CV algorithms averagely exhibit up to 2.98x performances speedups compared with their scalar counterparts. [ABSTRACT FROM AUTHOR]

Purpose: Mastoid abnormalities show different types of ear illnesses, however inadequacy of experts and low accuracy of diagnostic demand a new approach to detect these abnormalities and reduce human mistakes. The manual analysis of mastoid CT scans is time-consuming and labor-intensive. In this paper the first and robust deep learning-based approaches is introduced to diagnose mastoid abnormalities using a large database of CT images obtained in the clinical center with remarkable accuracy. Methods: In this paper, mastoid abnormalities are classified using the Xception based Convolutional Neural Network (CNN) model, with optimizer Adamax into five categories (Complete pneumatized, Opacification in pneumatization, Partial pneumatization, Opacification in partial pneumatization, None pneumatized). For this reason, a total of 24,800 slides of 152 patients were selected that include the mastoid from most upper to the lowest part of the middle ear cavity to complete the construction of the proposed deep neural network model. Results: The proposed model had the best accuracy of 87.80% (based on grader 1) and 88.44% (based on grader 2) on the 20th epoch and 87.70% (based on grader 1) and 87.56% (based on grader 2) on average and also significantly faster than other types of implemented architectures in terms of the computer running time (in seconds). The 99% confidence interval of the average accuracy was 0.012 which means that the true accuracy is 87.80% and 87.56% ± 1.2% that indicates the power of the model. Conclusions: The manual analysis of ear cavity CT scans is often time-consuming and prone to errors due to various inter- or intra operator variability studies. The proposed method can be used to automatically analyze the middle ear cavity to classify mastoid abnormalities, which is markedly faster than most types of models with the highest accuracy. [ABSTRACT FROM AUTHOR]

Massive Multi-Input Multi-Output (MIMO) and Coordinated MultiPoint (CoMP) technologies in Cloud-RAN (C-RAN) architecture become inevitable trend due to the advent of next-generation mobile applications, which are traffic-intensive, such as ultra high definition (UHD) video. In this paper, we study a joint beam activation and user scheduling problem in a 5G cellular network with massive MIMO and CoMP utilizing orthogonal random beamforming technique. This paper aims to minimize total Remote Radio Heads’ (RRHs’) energy expenditure in a dynamic C-RAN architecture while ensuring finite service time for all user traffic arrivals in the communication coverage. We leverage Lyapunov drift-plus-penalty framework to transform an original long-term average problem into a series of per-slot modified problems. Since the provided per-slot problem is combinatorial and nonlinear optimization problem, we are inspired by a greedy algorithm to design energy and delay guaranteed joint beam activation and user scheduling policy, namely BEANS. We prove that the proposed BEANS ensures finite upper bounds of average RRH energy consumption and average queue backlogs for all traffic arrival rates within constant ratio of capacity region and all energy-delay tradeoff parameters. These proofs are the first attempt to theoretically demonstrate guarantees of energy and queue bounds in a framework consisting of possibly negative submodular objective function and non-matriod constraints. Finally, via extensive simulations, we compare the capacity region and energy-queue backlog tradeoff of BEANS with optimal and existing algorithms, and show that BEANS attains up to 65% of energy saving for the same average queue backlog compared to the algorithms which do not take traffic dynamics and energy consumption into considerations. [ABSTRACT FROM AUTHOR]

The Model Predictive Control (MPC) framework has shown great potential for the control of Variable Speed Limits (VSLs) and Ramp Metering (RM) installations. However, the implementation to large freeway networks remains challenging. One major reason is that, by considering the VSLs to be discrete decision variables, an extremely difficult Mixed Integer Nonlinear Programming (MINLP) optimization problem has to be solved within every controller sampling interval. Consequently, many related papers relax the MINLP problems by considering the VSLs to be continuous variables. This paper proposes two novel MPC algorithms for coordinated control of discrete VSLs and continuous RM rates that do not make this relaxation. The proposed algorithms use a distributed control architecture and an alternating optimization scheme to relax the MINLP optimization problems but still consider the VSLs as discrete variables and, hence, offer a trade-off between computational complexity and system performance. The performance of the proposed algorithms is evaluated in a case study. The case study shows that relaxing the VSLs to be continuous variables with a distributed architecture results in a significant performance loss. Furthermore, both proposed algorithms have a lower computational complexity than the more conventional centralized approach and, as a result, they do manage to solve all optimization problems within the sampling intervals. Moreover, one of the proposed algorithms has a system performance that is remarkably similar to the optimal performance of the centralized approach. [ABSTRACT FROM AUTHOR]

With the flourishing development of deep learning (DL) and the convolution neural network (CNN), electroencephalogram-based (EEG) emotion recognition is occupying an increasingly crucial part in the field of brain-computer interface (BCI). However, currently employed architectures have mostly been designed manually by human experts, which is a time-consuming and labor-intensive process. In this paper, we proposed a novel neural architecture search (NAS) framework based on reinforcement learning (RL) for EEG-based emotion recognition, which can automatically design network architectures. The proposed NAS mainly contains three parts: search strategy, search space, and evaluation strategy. During the search process, a recurrent network (RNN) controller is used to select the optimal network structure in the search space. We trained the controller with RL to maximize the expected reward of the generated models on a validation set and force parameter sharing among the models. We evaluated the performance of NAS on the DEAP and DREAMER dataset. On the DEAP dataset, the average accuracies reached 97.94%, 97.74%, and 97.82% on arousal, valence, and dominance respectively. On the DREAMER dataset, average accuracies reached 96.62%, 96.29% and 96.61% on arousal, valence, and dominance, respectively. The experimental results demonstrated that the proposed NAS outperforms the state-of-the-art CNN-based methods. [ABSTRACT FROM AUTHOR]

This paper proposes a Software-Defined Delay Tolerant Networking (SDDTN) architecture as a solution to managing large Delay Tolerant Networking (DTN) networks in a scalable manner. This work is motivated by the planned deployments of large DTN networks on the Moon and beyond in deep space. Current space communication involves relatively few nodes and is heavily deterministic and scheduled, which will not be true in the future. It is unclear how these large space DTN networks, consisting of inherently intermittent links, will be able to adapt to dynamically changing network conditions. In addition to the proposed SDDTN architecture, this paper explores data plane programming and the Programming Protocol-Independent Packet Processors (P4) language as a possible method of implementing this SDDTN architecture, enumerates the challenges of this approach, and presents intermediate results. [ABSTRACT FROM AUTHOR]

Recently, deep learning has exhibited outstanding performance in various fields. Even though artificial intelligence achieves excellent performance, the amount of energy required for computations has increased with its development. Hence, the need for a new energy-efficient computer architecture has emerged, which further leads us to the neuromorphic computer. Although neuromorphic computing exhibits several advantages, such as low-power parallelism, it exhibits lower accuracy than deep learning. Therefore, the major challenge is to improve the accuracy while maintaining the neuromorphic computing-specific energy efficiency. In this paper, we propose a novel method of the inference process that considers the probability that after completing the learning process, a neuron can react to multiple target labels. Our proposed method can achieve improved accuracy while maintaining the hardware-friendly, low-power-parallel processing characteristics of a neuromorphic processor. Furthermore, this method converts the spike counts occurring in the learning process into probabilities. The inference process is conducted to implement the interaction between neurons by considering all the spikes that occur. The inferring circuit is expected to show a significant reduction in hardware cost and can afford an algorithm exhibiting a competitive computing performance. [ABSTRACT FROM AUTHOR]

Summary: Cache coherence in modern computer architectures enables easier programming by sharing data across multiple processors. Unfortunately, it can also limit scalability due to cache coherency traffic initiated by competing memory accesses. Rack‐scale systems introduce shared memory across a whole rack, but without inter‐node cache coherence. This poses memory management and concurrency control challenges for applications that must explicitly manage cache‐lines. To fully utilize rack‐scale systems for low‐latency and scalable computation, applications need to maintain cached memory accesses in spite of non‐coherency. This paper introduces Bounded Incoherence, a memory consistency model that enables cached access to shared data‐structures in non‐cache‐coherency memory. It ensures that updates to memory on one node are visible within at most a bounded amount of time on all other nodes. We evaluate this memory model on modified PowerGraph graph processing framework, and boost its performance by 30% with eight sockets by enabling cached‐access to data‐structures. [ABSTRACT FROM AUTHOR]