Search

Abstract: Planning is motivated by acting. Most of the existing work on automated planning underestimates the reasoning and deliberation needed for acting; it is instead biased towards path-finding methods in a compactly specified state-transition system. Researchers in this AI field have developed many planners, but very few actors. We believe this is one of the main causes of the relatively low deployment of automated planning applications. In this paper, we advocate a change in focus to actors as the primary topic of investigation. Actors are not mere plan executors: they may use planning and other deliberation tools, before and during acting. This change in focus entails two interconnected principles: a hierarchical structure to integrate the actorʼs deliberation functions, and continual online planning and reasoning throughout the acting process. In the paper, we discuss open problems and research directions toward that objective in knowledge representations, model acquisition and verification, synthesis and refinement, monitoring, goal reasoning, and integration. [Copyright &y& Elsevier]

Abstract: Two categories of mathematical models were compared for the simulation of consolidation of paper coating structures, that is for the packing of pigments on a paper substrate under dewatering conditions. The first category uses probabilistic methods, relying on a random number generator to either determine the initial position of the pigments or their motion. The second category uses deterministic methods based on force balances. In this work, two probabilistic models and two deterministic models are described and their respective advantages and drawbacks are critically reviewed. Simulation results obtained using three of these methods are compared for the case of monodisperse and bidisperse spherical suspensions. Porosity calculations of the numerical packings obtained with the (deterministic) discrete element method (DEM) and two probabilistic methods, the Monte-Carlo (MCD) and the steepest descent (SDD) deposition methods, are compared with experimental data from the literature. These calculations reveal significant differences in the pore volume obtained with these three models. An analysis based on the bridging and relaxation phenomena that prevail in the flow of such particulate systems provide an explanation for these differences and show the strong potential of the discrete element method. The choice of the simulation method depends on the objective of the simulations. DEM will provide more accurate predictions of macroscopic quantities such as the porosity or the roughness, but requires very long computational times. MCD or SDD will only provide qualitative trends, but is computationally far less intense. A combination of strategies might be appropriate, using MCD (or SDD) to provide guidelines and DEM to enhance the results predicted by MCD. [Copyright &y& Elsevier]

Abstract: The lack of equivalence between computerized and pencil-and-paper administration in measures of negative affect have been attributed to variance created by negative affect towards computers or computer anxiety (CA). In the current study, paper baseline computer anxiety and state/trait anxiety measures were obtained from 51 first-year psychology undergraduate volunteers. Further measures were taken by either paper or computer before and after students received their grade for the first research methods practical report of the course. Levels of state anxiety (SA) were found to increase significantly at this time. A lexical decision task was completed at each measurement stage as an additional behavioural mood indicator. Results revealed that CA was only related to SA prior to receipt of grade, at a point where equivalence between administration method of measures had been demonstrated. Non-equivalence in measurement of SA occurred after students had received their mark, in that levels of anxiety increased in the computer condition and decreased in the paper condition. Lack of equivalence, therefore, appears to be a function of psychological stress, characterised by affective modulation rather than CA. [Copyright &y& Elsevier]

Abstract: The present study was performed to explore a new approach to systematization of the management of paper-based clinical pathways by developing a new system requiring little capital investment. A new system was developed and incorporated into an existing network at a hospital with a paper-based clinical pathway management system. The effectiveness of this new system was examined by comparing the management efficiency of clinical pathways before and after its introduction, and by comparison of the new system with other such systems currently in place at other medical institutions with regard to efficiency. In addition, the acceptability of the system for other medical institutions was examined by providing free access to the software on the Internet. The development costs of the new system were low. Although the new system has been in place for more than 3 years, no problems have yet been encountered in either the existing network system or in the management system itself. The new system allows the processing of statistics and analysis of circulation or variance automatically, neither of which were possible in the original paper-based system. We provided open access to the system as free software on the Internet, and it has since been downloaded by many medical institutions and enterprises in Japan. This system is very useful for institutions where it is difficult to introduce expensive new systems for systematic management of clinical pathways, such as electronic medical records, because of problems regarding capital or system management, and it may also be useful in other countries. [Copyright &y& Elsevier]

One of the most powerful functional evaluations of the electrical activity of the brain is the EEG imaging, but wide clinical use is limited by its costs. It is also of clinical, academic and scientific interest to obtain brain electrical maps from old paper/ink, patient recordings. The aim of the present study was the development of a computer system designed to obtain bidimensional and tridimensional maps, continuous movie map display and mosaic presentation from conventional paper/ink EEG recordings. The wave amplitude was manually measured with a translucent template from conventional 8- to 16-channel EEG paper recordings using 10–20 monopolar montage for one or both hemispheres. The computer system allows the selection of the number and location of electrodes, input of amplitude values, and the map display mode. The electrical brain field was generated from amplitude measurements by a spherical splines interpolation algorithm on a conventional Pentium®-based computer. The interpolated surface was represented on a semi-sphere modeled skull. The EEG maps displayed with pseudo-color or gray scales can be rotated, zoomed in or zoomed out and/or printed for clinical reports. Movie animation or mosaic display of space-temporal EEG voltage changes were generated by processing sequential amplitude measurements. This system represents a cost-effective method for EEG mapping from conventional paper/Ink EEG equipments. [Copyright &y& Elsevier]

This article presents information about the special issue of "FLOSS at Large: Selected Papers from the FLOSS Workshop at 4SEASST Joint Conference, Paris 25-28 August 2004," which focuses on the many strands of the Free/Libre Open Source Software (FLOSS) movement.

Adversarial attacks can mislead deep learning models to make false predictions by implanting small perturbations to the original input that are imperceptible to the human eye, which poses a huge security threat to computer vision systems based on deep learning. Physical adversarial attacks, which is more realistic, as the perturbation is introduced to the input before it is captured and converted to a image inside the vision system, when compared to digital adversarial attacks. In this paper, we focus on physical adversarial attacks and further classify them into invasive and non-invasive. Optical-based physical adversarial attack techniques (e.g. using light irradiation) belong to the non-invasive category. The perturbations can be easily ignored by humans as the perturbations are very similar to the effects generated by a natural environment in the real world. With high invisibility and executability, optical-based physical adversarial attacks can pose a significant or even lethal threat to real systems. This paper focuses on optical-based physical adversarial attack techniques for computer vision systems, with emphasis on the introduction and discussion of optical-based physical adversarial attack techniques. [ABSTRACT FROM AUTHOR]

In this paper, the compound Poisson phase-type process is defined and analyzed. This paper proves that for a non-negative compound Poisson phase-type process, the compound value for all the arrivals by a given time can be approximated by a phase-type distribution. As an application of this process, three different shock models are studied: the cumulative shock model, a degradation-threshold-shock model, and a shock model for the multi-component system. A novel approach is proposed to compute the system's reliability under the aforementioned shock models for a general counting process. Numerical illustrations are presented. [ABSTRACT FROM AUTHOR]

• A micro-measurement method based on the IPDA. This paper uses a low-pass filter to process the zero-mode current to obtain the characteristic components. Then, by calculating the second derivative of the filtered part, the selection of the inflection point dense area (IPDA) at the microscopic level is achieved. This physical significance of IPDA is used to effectively express the feature that the zero-mode current direction of the faulty feeder is entirely opposite to that of the healthy. • It is proposing an ultra-high-speed protection scheme for DC distribution networks. In this paper, only 20 sampling points within the IPDA are processed, and the selection of the data window length can be adjusted according to the system sampling frequency and other factors. It meets the requirements of DC distribution networks for protection speed and selectivity, and the algorithm can correctly detect faulty feeders in a high resistance state, with little influence from external factors. A flexible direct current (DC) distribution network based on modular multilevel converter (MMC), with the advantages of no phase change failure and fewer current conversion links. It has gradually become the development trend of future medium voltage distribution networks. However, when a single-pole ground fault occurs in the feeder, the converter's low inertia and weak damping characteristics lead to a fast-rising fault current with a high peak value, which is harmful to the system. In this paper, for the difficulty of accurately selecting the faulty feeder when a single-pole ground fault occurs in a radial DC distribution network, a detection method based on the zero-mode current within the inflection point dense area (IPDA) for the fluctuation characteristics of concave and convexity (FCCC) is proposed. Firstly, the zero-mode current of each feeder is processed using a low-pass filter to select the eigen-components; Then, taking second order derivatives of the eigen-components, and the IPDA is established, to find a period in micro measurement to achieve accurate direction correspondence with opposite directions; Again, the data within IPDA is normalized; the criterion is constructed by determining each feeder's FCCC. Finally, the simulation model is built by Power Systems Computer Aided Design to verify that the method can identify the faulty feeder in various situations. Compared with existing methods, the fault detection method proposed in this paper has good stability and can withstand large grounding resistances. [ABSTRACT FROM AUTHOR]

Process development is typically associated with lengthy wet-lab experiments for the identification of good candidate setups and operating conditions. In this paper, we present the key features of a model-based approach for the identification and assessment of process design space (DSp), integrating the analysis of process performance and flexibility. The presented approach comprises three main steps: (1) model development & problem formulation, (2) DSp identification, and (3) DSp analysis. We demonstrate how such an approach can be used for the identification of acceptable operating spaces that enable the assessment of different operating points and quantification of process flexibility. The above steps are demonstrated on Protein A chromatographic purification of antibody-based therapeutics used in biopharmaceutical manufacturing. • Model-based approach for quantitative design space characterization. • Integrated assessment of design decisions, operating variables and process variabilities. • Application to chromatographic unit operation. [ABSTRACT FROM AUTHOR]

This paper considers the data-driven fault isolation and estimation problem for linear time-invariant systems with unknown dynamic matrices and multiple actuator faults. In most of existing fault isolation methods, how to accurately identify the types of faults has not been solved well when the system matrices are unknown. To deal with this problem, a neural network-based fault isolation method is proposed by analyzing and extracting features of different fault models in terms of constructing sparse vectors and function libraries using the available input–output data. Then, a fault estimator is designed to estimate the fault signals within the data-driven framework, where its parameters are computed by the system's Markov parameters and the identified types of faults. Finally, two examples are used to verify the advantages and effectiveness of the proposed fault isolation and estimation approach. • A data-driven fault diagnosis approach is developed for unknown linear systems. • A neural network-based fault isolation method is used to isolate actuator faults. • The faults features of different fault models are analyzed and extracted. • A fault estimator is designed for both single and simultaneously occurring faults. • H 2 index is used to attenuate the effects of noises on fault estimation error. [ABSTRACT FROM AUTHOR]

This paper focuses on the classification of classes of topological equivalence of finite group actions on Riemann surfaces. By the Riemann-Hurwitz bound, there are just finitely many groups that act conformally on a closed orientable surface S g of genus g ≥ 2. With each such action of a group G on S g one can associate the fundamental group Γ = π (O) of the quotient orbifold O = S g / G , isomorphic to a Fuchsian group determined completely by orbifold's signature. The Riemann existence theorem reduces the problem of the existence of an action of G on S g to a purely group-theoretical problem of deciding whether there is an smooth epimorphism mapping the Fuchsian group Γ onto the group G. Using computer algebra systems such as Magma or GAP, together with the library of small groups, the generation of all finite group actions on a surface of fixed small genus g ≥ 2 becomes almost a routine procedure. The difficult part is to determine the classes of these actions with respect to topological equivalence. To achieve this, one needs to investigate the action of the automorphism group of a Fuchsian group on the set of finite group actions on S g with the corresponding signature. In this paper we derive several results on the topological equivalence of finite group actions on Riemann surfaces. As an application, we derive complete lists of finite group actions of genus g ≤ 8 distinguished up to the topological equivalence. [ABSTRACT FROM AUTHOR]

• A recognized microwave waveform generation is proposed using time-delayed feedback reservoir computing. • This work demonstrates that an SOA-based time-delayed feedback reservoir computing is constructed and theoretically analyzed. • A variety of optimization algorithms are used to demonstrate and compared the automated capability of waveform synthesis. • The paper demonstrates the six waveforms with low RMSE generated by the proposed microwave waveform generation. This paper proposes and demonstrates a model for generating microwave waveforms using a delay feedback photonic reservoir computing system with SOA. Our scheme involves building a photonic reservoir computer with SOA, which provides rich nonlinearity functionalities and memory capability for the reservoir. The automatic optimization of output waveform is implemented through the incorporation of gradient descent regression and a reservoir output layer. In theory, the system is capable of generating arbitrary microwave waveforms towards the target automatically. In our work, the capacity of the model is tested through theoretical analysis and simulation. We also compared the optimization effects of several regression algorithms applied to the reservoir computing, and generated six typical microwave waveforms, including triangle, square, sawtooth, negative, oblique triangle 1, and oblique triangle 2. [ABSTRACT FROM AUTHOR]

This paper evaluates the dimensional accuracy of industrial X-ray computed tomography (CT) measurements as a function of the geometric magnification (M g) used for CT scanning. Two plates of square-shaped geometry with 28 drilled holes, one made of aluminum (48 mm × 48 mm × 8 mm) and a second one made of steel (6 mm × 6 mm × 1 mm), are used as measuring workpieces to test fourteen different magnification axis positions in a metrology-grade X-ray CT system. In addition to measuring the diameter and roundness of each hole, distance measurements between the holes are evaluated—both as uni-directional (center-to-center) and bi-directional (point-to-point or edge-to-edge) lengths. The variance of CT dimensional measurements, with respect to reference data obtained from tactile coordinate measurement machines (CMMs), is studied. It is found that the measurement deviations for uni-directional distances are approximately independent of M g and mostly within a ± 4 μm range. In contrast, for measurements of diameter, roundness, and bi-directional lengths, the deviations between CT and CMM data extend outside the ± 4 μm limits and strongly depend on M g (which is a variable inversely proportional to the voxel size of CT scan, V X). More specifically, decreasing M g or, equivalently, increasing V X , generally leads to larger deviations in the dimensional data. To include the magnification dependence on the estimations of CT measurement deviations, based on the experimental data presented throughout this paper, the authors propose that the maximum error of CT length measurements can be expressed as E m a x = (A + X / B + L / K) μm, with A , B , and K being constant factors (determined for a particular CT measuring setup), L the dimension of the measuring length (in mm), and X the magnification axis position of the CT scanner (also in mm). Whereas the term " L / K " considers error influences from the size of the sample, the term " X / B " accounts for influences from magnification (and voxel size) selections during CT measurement. • X-ray computed tomography (CT) is studied via changes of geometric magnification M g. • Errors of uni-directional center-to-center distances are approx. independent of M g. • Errors of form and bi-directional lengths strongly depend on M g (and voxel size V X). • Dimensional accuracy of CT data deteriorates as M g lowers and the V X grows. • A model that includes M g is proposed for estimating maximum CT measurement errors. [ABSTRACT FROM AUTHOR]

X-ray computed tomography (X-ray CT) is a new coordinate measuring technology that can measure both the external and internal features. However, its measuring accuracy is insufficient compared with coordinate measuring machines. To improve the measurement accuracy of the X-ray CT system, it is necessary to evaluate and compensate for each error component that influences the dimensional measurement results. In this paper, we propose a method for evaluating and compensating for the geometrical errors of the X-ray CT system, which is one of the dominant error factors, using a laser tracking interferometer. A laser tracking interferometer is not required the re-alignment of the laser path, and the data can be obtained using a common coordinate system. The proposed method was applied to a CT system designed for non-destructive testing. The result clarified that the geometrical errors of the X-ray CT system were over 100 μm. Additionally, the errors in measuring the distance between the spheres were demonstrated to be reduced by compensating for the geometrical errors of the CT system. • The method for evaluating and compensating the geometrical errors of an X-ray CT system using a laser tracker was proposed. • The experimental results showed that the compensation method reduced the scale error of the X-ray CT volume data. • The applicability of the proposed method is limited by the CT system to the measurement accuracy of the laser tracker. [ABSTRACT FROM AUTHOR]

This paper presents a novel approach based on multi-agent reinforcement learning for spacecraft formation flying reconfiguration tracking problems. In this scheme, spacecrafts learn the control strategy via transfer learning. For this matter, a new generalized discounted value function is introduced for the tracking problems. Due to the digital nature of spacecraft computer systems, local optimal controllers are developed for the spacecrafts in discrete-time. The stability of the controller is proven. Two Q-learning algorithms are proposed, in each of which the optimal control solution is learned on-line without knowledge about the system dynamics. In the first algorithm, each agent learns the optimal control independently. In the second one, each agent shares the learned information with other agents. Next, the collision avoidance capability is provided. The effectiveness of the presented schemes is verified through simulations and compared with each other. [ABSTRACT FROM AUTHOR]

Recently, smart data has attracted great attention in the smart city community since it can provide valuable information to support intelligent services such as planning, monitoring, and decision making. However, it imposes a big challenge to explore smart data from big data gathered from smart city with various advanced fusion and analysis approaches. This paper proposes an incremental tensor-based fuzzy c-means approach (IT-FCM) for obtaining smart data from continuously generated big data. Specifically, a weighted version of the tensor-based fuzzy c-means approach (T-FCM) is firstly proposed to cluster the dataset that combines the previous cluster centroids and the new generated data. Aiming to improve the clustering efficiency, the old data objects are represented by the centroids to avoid repeat clustering. Furthermore, this paper presents an edge-cloud-aided clustering scheme to fuse big data from different sources and perspectives and further to implement co-clustering on the fused datasets for exploring smart data. Finally, the proposed IT-FCM approach is evaluated by comparing with T-FCM regarding clustering accuracy and efficiency on two different datasets in the experiments. The results state that IT-FCM outperforms T-FCM in clustering streaming big data in terms of accuracy and efficiency for obtaining smart data. • An edge-cloud-aided computing system for co-clustering is presented. • An edge-cloud-aided computing system for co-clustering is presented. • Experiments are conducted to evaluate the presented approach. [ABSTRACT FROM AUTHOR]

Applicative computing systems and technologies have taken a strong position in modern computing. In this paper, the basic applicative system, whether it is a lambda calculus or a system of combinators, is considered as a prototype concept system, using which it is possible to build individual systems that are practically significant for mathematics, computing, or programming. They are families of computational models that have both their own semantics and applied areas. This conceptualization/individualization technique is characteristic of the field of semantic studies. As it turns out, the applicative approach forms a metatheoretical framework that provides the basis for cognitive systems that consider abstract objects and interpret their properties and behavior in the environment of modern computing. [ABSTRACT FROM AUTHOR]

Haze usually causes severe interference to image visibility. Such degradation on images troubles both human observers and computer vision systems. To seek high-quality images from degraded ones, a large number of image dehazing algorithms have been proposed from different perspectives like image enhancement, restoration, and fusion. Especially in recent years, with the rapid development of deep learning, CNN-based methods have already dominated the mainstream of image dehazing and gained significant progress on benchmark datasets. This paper firstly presents a comprehensive survey of existing image dehazing methods, and then conducts both qualitative and quantitative comparisons among representative methods, from classic methods to recent advanced approaches. We expect the literature survey and benchmark analysis could help readers better understand the advantages and limitations of existing dehazing methods. Moreover, a discussion on possible trends in single image dehazing is put forward to innovate further works. • Image dehazing methods via enhancement, restoration and fusion are surveyed. • The datasets of image dehazing and widely used metrics are summarized and discussed. • Experiments of 19 methods over 8 metrics are conducted to evaluate the performance. • Discussion of opening problems and future trends are given to innovate future works. [ABSTRACT FROM AUTHOR]

Residual connections have been established as a staple for modern deep learning architectures. Most of their applications are cast towards feedforward computing. In this paper, we study the architectural bias of residual connections in the context of recurrent neural networks (RNNs), specifically in the temporal dimension. We frame our discussion from the perspective of Reservoir Computing and dynamical system theory, focusing on important aspects of neural computation like memory capacity, long-term information processing, stability, and nonlinear computation capability. Experiments corroborate the striking advantage brought by temporal residual connections for a plethora of different time series processing tasks, comprehending memory-based, forecasting, and classification problems. • We study residual (skip) connections in the context of RNNs in the temporal dimension. • Temporal residual connections enable long-term processing of time series. • Orthogonal skip connections allow to drive RNNs to the edge-of-stability in an approximate dynamical isometry regime. • Experiments on memory, forecasting, and classification tasks show the benefits of temporal residual connections in RNNs. [ABSTRACT FROM AUTHOR]

Real-time virtualization is currently a hot topic, and there is much ongoing research on real-time Virtual Machines and hypervisors. However, most of the previous research focused either on reducing the latencies introduced by the virtualization stack (hypervisor, host Operating System, Virtual Machine scheduling, etc...) or analyzing the virtual CPU scheduling algorithms. Only a few works investigated the impact of the guest Operating System architecture on real-time performance or considered multiple performance metrics (latency, schedulability, startup times, resource consumption) at the same time. This paper compares various features of different virtualization technologies and guest Operating Systems, evaluating their suitability for serving real-time applications. The results indicate that solutions based on KVM (and an appropriate microvm) and the OSv unikernel can be considered viable alternatives to more traditional VMs or containers. [ABSTRACT FROM AUTHOR]

Modern edge real-time automotive applications are becoming more complex, dynamic, and distributed, moving away from conventional static operating environments to support advanced driving assistance and autonomous driving functionalities. This shift necessitates formulating more complex task models to represent the evolving nature of these applications aptly. Modeling of real-time automotive systems is typically performed leveraging Architectural Languages (ALs) such as Amalthea, which are commonly used by the industry to describe the characteristics of processing platforms, operating systems, and tasks. However, these architectural languages are originally derived for classical automotive applications and need to evolve to meet the needs of next-generation applications. This paper proposes an automatic framework for the modeling and automatic code generation of dynamic automotive applications under the QNX RTOS. To this end, we extend Amalthea to describe chains of communicating tasks with multiple operating modes and to consider the QNX's reservation-based scheduler, called APS, which allows providing temporal isolation between applications co-located on the same hardware platform. Finally, an evaluation is presented to compare different implementation alternatives under QNX that are automatically generated by our code generation framework. [ABSTRACT FROM AUTHOR]

• A multifunctional on-chip optical computing system (decoder half-adder and multiplier) is presented. • Firstly achieve an optical form of traditional digital multiplier by directed logic. • A compact optical design utilized only six MRRs for three computing functionalities. • Provide new ideas for the design of optical computing devices. Photonic integrated circuits are suffering rapid development and attracting lots of interests in various application fields, however, obstacles like low integration of photonics components and single functionality are severely hindering the progress. In this paper, we proposed a compact on-chip optical computing system, based on directed logic and micro-ring resonators (MRRs), achieving multiple functionalities including optical decoder, half-adder and multiplier. Through clever design, that proposed merely use several bus waveguides and six MRRs, and resulting in compactness with a footprint of 46.45 μ m × 35.44 μ m. Simulation results by finite-difference time-domain (FDTD) demonstrate the effectiveness and feasibility of the proposed multifunctional optical computing system. [ABSTRACT FROM AUTHOR]

Asymmetric Multicore Processors (AMP) are widely used in both advanced and basic computing systems due to their fundamental flexibility and outstanding computing possibilities. Among the subcategories of AMP, Performance Asymmetric Multicore (PAM) Architectures are unique in that they include various micro-architecture cores into a single chip. Nevertheless, the complex interplay between heterogeneous cores and a diverse range of applications presents a formidable challenge in determining the optimal hardware configuration, encompassing core selection and reduced energy consumption for each application. To tackle these multifaceted problems, this paper introduces a pioneering model for core prediction and resource allocation based on the Dual Branch Recurrent Neural Network (DB-RNN), specifically for AMP. The DB-RNN model encompasses weight sharing facilitated by a hybrid optimization algorithm named African Vulture with Aquila Optimizer (AVAO), and a core prediction module. The proposed model evaluates the Energy-Delay Product (EDP) to find the performance of each core. In the final phase, DB-RNN dynamically predicts the most suitable cores for individual workloads at runtime, thereby elevating both energy efficiency and overall system performance. The experimental results show that the model achieved prediction accuracy up to 98.76 %. This innovative approach paves the way for enhanced efficiency and performance in PAM systems. [ABSTRACT FROM AUTHOR]

A physical system is determined by a finite set of initial conditions and "laws" represented by equations. The system is computable if we can solve the equations in all instances using a "finite body of mathematical knowledge". In this case, if the laws of the system can be coded into a computer program, then given the initial conditions of the system, one can compute the system's evolution. Are there incomputable physical systems? This question has been theoretically studied in the last 30–40 years. In this paper, we experimentally show for the first time the strong incomputability of a quantum experiment, namely the outputs of a quantum random number generator. Moreover, the experimental results are robust and statistically significant. [ABSTRACT FROM AUTHOR]

Reservoir Computing (RC), an evolution from Recurrent Neural Networks (RNN), not only represents a unique machine learning paradigm, but also serves as a neuromorphic framework that mirrors the intricate cortical circuits of the human brain. This paper proposes another new photonic RC system based on four basic photonic reservoir computing architectures (single photonic RC system, the parallel photonic RC system, the dual-feedback loop-based photonic RC system and the mutually coupled photonic RC system). System proposed uses optical injection for signal input and retains two parallel responsive semiconductor lasers (R -SLs) with self-feedback loops. Meanwhile, two relatively independent R -SLs are mutually coupled via two coupling lines. The new photonic RC system adds only two sections of fiber compared to the parallel photonic RC system and the mutually coupled photonic RC system. The experiments show that the system proposed has significant advantages on the nonlinear auto regressive moving average series tasks, the chaotic time series prediction tasks and the waveform classification task. More importantly, the memory capacity of system proposed can be adjust by controlling the delay time of the self-feedback loops, so it has higher memory capacity to handle the higher order nonlinear auto regressive moving average tasks (NARMA20 and NARMA30) after optimizing the parameters. • The proposed system integrates the advantages of the dual-feedback loop-based and mutually coupled photonic RC systems. • The proposed system can adjust its memory capacity to suit various short-term memory needs. • The inclusion of mutual coupling and self-feedback structures in the proposed system enriches its dynamic characterization. [ABSTRACT FROM AUTHOR]

• Facilitates advanced Weigh-In-Motion (WIM) techniques for pavement management plan. • Hybrid system with GFRP-FBG sensors with computer vision for WIM measurements. • Addresses wander effect challenge in vehicle weight monitoring. • Provides novel insights into vehicle–road interaction dynamics. The paper addressed the challenges of the two main weigh-in-motion (WIM) systems through a hybrid WIM system, which combines embedded in-pavement sensor systems and computer vision to enhance accuracy while minimizing human intervention. To establish a cost-effective system, this study employed a standard camera integrated with computer vision technique to improve the precision of identifying wheel location and automatically adjust distance calibration, particularly in response to variations in camera angle. In addition, the system achieved real-time vehicle weight detection efficiently by addressing the vehicle wander effect concerns of the in-pavement sensors through camera using computer vision. The system can achieve weight accuracy notably exceeding ASTM 1318E-09 standards, with over 90% accuracy for distances below 0.1 m, demonstrating its potential value in assisting traffic analysis and pavement maintenance by offering real-time, accurate WIM measurements at a low cost for instrumentation. [ABSTRACT FROM AUTHOR]

Imaging spectroscopy integrates traditional computer vision and spectroscopy into a single system and has gained widespread acceptance as a non-destructive scientific instrument for a wide range of applications. The current state of imaging spectroscopy spans diverse applications including but not limited to air-borne and ground-based computer vision systems. This paper presents the current state of research and industrial applications including precision agriculture, material classification, medical science, forensic science, face recognition and document image analysis, environment monitoring, and remote sensing, which can be aided through imaging spectroscopy. In this regard, we further discuss a comprehensive list of applications of imaging spectroscopy, pre-processing techniques, and spectral image acquisition systems. Likewise, publicly available databases and current software tools for spectral data analysis are also documented in this review. This review paper, therefore, could potentially serve as a reference and roadmap for people looking for literature, databases, applications, and tools to undertake additional research in imaging spectroscopy. • A survey on the current state of imaging spectroscopy. • Pre-processing techniques and data acquisition methods are highlighted. • We provide applications of imaging spectroscopy in various domains. • Publicly available databases and software tools for spectral data are presented. [ABSTRACT FROM AUTHOR]

The rapid informatization and digitalization of the society heavily rely on the extensive use of parallel and distributed, networked computer systems. It is important for large-scale parallel and distributed systems to be able to detect and tolerate faulty vertices in the network. A network's fault status can often be characterized with the network's connectivity and diagnosability. The connectivity/diagnosability can be defined under various conditions. This paper is concerned with the connectivity/diagnosability under the " g -good-neighbor condition", which can more accurately measure a network's fault status. In this paper, we propose a new class of recursive networks, named recursive match networks (RMNs), which contain the well-known BCube and BC networks. We determine the RMNs' g -good-neighbor connectivity and g -good-neighbor conditional diagnosability under the classic MM* and PMC diagnostic models for g ≥ 0. Since the RMN is a more general network covering the BCube and BC networks, our results can be directly applied to these two networks. • We proposed a new class of networks, named recursive match networks (RMNs), which contain the BCube, BC networks, and some other networks. • We obtained the g -good-neighbor connectivity and g -good-neighbor conditional diagnosability of RMNs, under the MM* and PMC models. • We got that the g -good-neighbor connectivity/diagnosability of the RMN are both nearly g + 1 times of its original, unconditional ones. • The results of this paper can be directly applied to the BCube, BC networks, and any other networks that fall within the definition of the RMN. [ABSTRACT FROM AUTHOR]

Deploying deep learning models in time-critical applications with limited computational resources, for instance in edge computing systems and IoT networks, is a challenging task that often relies on dynamic inference methods such as early exiting. In this paper, we introduce a novel architecture for early exiting based on the vision transformer architecture, as well as a fine-tuning strategy that significantly increase the accuracy of early exit branches compared to conventional approaches while introducing less overhead. Through extensive experiments on image and audio classification as well as audiovisual crowd counting, we show that our method works for both classification and regression problems, and in both single- and multi-modal settings. Additionally, we introduce a novel method for integrating audio and visual modalities within early exits in audiovisual data analysis, that can lead to a more fine-grained dynamic inference. • We propose a more accurate architecture with less overhead for early exit branches. • Our method achieves higher accuracy due to a global receptive field. • Our method is general purpose, works in multi-modal settings and for several problems. • We introduce a novel method for integrating audio and visual features in early exits. • We introduce a fine-tuning strategy for early exits. [ABSTRACT FROM AUTHOR]

In this paper numerical approximations are computed for the solution of a system of two reaction-convection-diffusion equations by a fitted mesh finite difference method. The method suggested is established to have almost first order parameter robust convergence and is illustrated through examples corroborating the theory. [ABSTRACT FROM AUTHOR]

Energy efficiency has emerged as one of the most essential design parameters in contemporary computing system design. Approximate computing is a new computing paradigm to achieve energy efficiency by trading-off energy/area/latency improvements with accuracy for error-resilient applications. This paper proposes R econfigurable E nergy-efficient A pproximate D ivider (READ) that achieves several energy–quality configurable modes using fixed restoring array divider architecture. Conventional approximate binary dividers require various divider hardware configurations to achieve distinct energy–quality trade-off points, which decreases the hardware flexibility, especially for modern embedded systems. READ accomplishes energy efficiency while meeting the dynamically varying accuracy requirements of the targeted application. READ uses reconfigurable subtractor cells that can work in either accurate or approximate mode using a subtractor cell controller logic. The paper also introduces the design of overflow detector using minimal hardware resources. A comprehensive accuracy and hardware evaluation on CMOS 45-nm technology node are performed for the proposed dividers as well as other state-of-the- art divider designs. Compared to the accurate 16 ∕ 8 divider design, the proposed divider shows an improvement of 49% in terms of energy efficiency and is 1.26x faster, while introducing minimal errors. The proposed divider design is demonstrated for its efficacy in image processing tasks and shows nominal effect on the output quality. • A fixed restoring array based approximate divider is proposed and presented with detailed error and circuit analysis for CMOS-45 nm technology node. • The approximate divider presented is reconfigurable in runtime which enables accuracy configurability by varying the design parameter. • Reconfigurable half and full subtractor cells are designed and proposed which achieves dual mode of operation — accurate and exact with energy savings. • Overflow detector circuit for divider design is proposed and presented which is scalable and hardware efficient. • The approximate divider designs are evaluated for image processing applications and found to be satisfactory in terms of Quality of Services (QoS). [ABSTRACT FROM AUTHOR]

Minimizing energy consumption is a critical challenge for real-time workflows, particularly in heterogeneous cloud computing systems. State-of-the-art algorithms aim to minimize the energy consumed for processing such applications by choosing virtual machines (VMs) to shut down from all opened VMs (i.e., VM merging). However, such VM merging through an "on-to-close" approach usually incurs high computational complexity. This paper proposes an energy-efficient VM opening (EEVO) algorithm that is capable of choosing VMs to turn on from all closed VMs while satisfying the real-time constraint of applications. Considering that there are slacks that can be eliminated or reduced between adjacently scheduled tasks after using the EEVO algorithm, a dynamic scaling down EEVO algorithm (DEEVO) is further proposed. DEEVO is implemented by scaling down the frequency of VMs executing each task based on the dynamic voltage and frequency scaling (DVFS) technique. Experimental results demonstrate that, with the above-mentioned improvements, DEEVO achieves lower energy consumption for real-time workflows than state-of-the-art algorithms do. In addition, DEEVO outperforms state-of-the-art algorithms in the computational efficiency of accomplishing task scheduling. [ABSTRACT FROM AUTHOR]

Cloud computing leads an organization for data storage, sharing, processing, and other services. It was subjected to several challenges insecurity due to the presence of regular attacks. These security challenges are worsened due to the presence of various attack environments. The conventional techniques adopted in cloud security are Intrusion Detection System (IDS). However, the IDS system requires an efficient security model for improving security in the cloud. In this paper, to improve the security in the cloud IDS system developed a framework stated as IDSGT-DNN. The proposed model incorporates an attacker and a defender mechanism for attack and normal data processing. The developed game theory is implemented in the DNN model with IWA for the identification of optimal solutions. The estimation is based on the identification of minimal and maximal optimal points in the dataset. The proposed IDSGT-DNN model is evaluated for CICIDS - 2017 dataset. The performance of proposed IDSGT-DNN is evaluated with existing technique is performed. Simulation analysis exhibited that proposed IDSGT-DNN exhibits enhanced performance in terms of accuracy, detection rate, and precision, F-Score, AUC, and FPR. [ABSTRACT FROM AUTHOR]

• A new model of electron emission processes in strong electromagnetic fields is formulated. • The model takes into account the structure, properties and actual geometry of the emitter. • The factors (aperture, frequency range, direction diagrams) influencing the parameters of the electron beams are calculated. • The numerical procedure combines methods: control volumes, particle-in-cells and large smoothed particles. • The parallel implementation integrates domain decomposition, parallelization on particles and load balancing. The aim of this work is modeling processes of field electron emission in strong electromagnetic fields. This problem is relevant for many technical and medical applications. At present time, electrical devices that combine a large value of field, a powerful relativistic effect and an ultra-short time interval of action are in demand. They find their application in the treatment of the surfaces with inorganic, organic and mixed structures. Modeling of such devices encounters certain difficulties due to the complexity of the mathematical description of the emission processes. In this paper, an approach using the method of large smoothed particles in combination with grid calculation of fields based on Maxwell's equations is proposed. The study was carried out within the framework of the problem of calculating the field emission of electrons from the surface of axisymmetric metal cathodes on Cartesian and unstructured curved meshes. To implement the approach, a complex mathematical model, a parallel numerical algorithm and its software realization have been developed. The elaborated software is focused on the use of multiprocessor computing systems with a central architecture. Test calculations confirmed the correctness of the proposed approach and the high efficiency of its software implementation. [ABSTRACT FROM AUTHOR]

Lie groups of symmetries of differential equations constitute a fundamental tool for constructing group-invariant solutions. The number of subgroups is potentially infinite and so the number of invariant solutions; thus, it is crucial to obtain a classification of subgroups in order to have an optimal system of inequivalent solutions from which all other solutions can be derived by action of the group itself. Since Lie groups are intimately connected to Lie algebras, a classification of inequivalent subgroups induces a classification of inequivalent Lie subalgebras, and vice versa. A general method for classifying the Lie subalgebras of a finite–dimensional Lie algebra uses inner automorphisms that are obtained by exponentiating the adjoint groups. In this paper, we present an effective algorithm able to automatically determine optimal systems of Lie subalgebras of a generic finite–dimensional Lie algebra abstractly assigned by means of its structure constants, or realized in terms of matrices or vector fields, or defined by a basis and the set of non-zero Lie brackets. The algorithm is implemented in the computer algebra system Wolfram Mathematica ™; some meaningful and non-trivial examples are considered. [ABSTRACT FROM AUTHOR]

Amplitude–wavelength (AW) maps or "Stedman diagrams" are often used to provide a graphical representation of the limitations and capabilities of surface measuring instruments. This paper presents an approach for setting the parameter constraints of X-ray computed tomography (CT) in terms of resolution and measuring range for the purpose of representing the performance of industrial CT systems on an AW map. Such AW map will allow instrument users to quickly compare the CT instrument performance to other measuring systems. Examples of the construction of AW maps for different working capabilities of X-ray CT systems, and based on experimental data, are provided. Polypropylene, aluminum, and steel are three workpiece materials considered for determining some of the limitations of measuring capability for the maps developed in this paper. Although the limiting boundaries of the AW plots presented in this paper set a generalized measure of performance for comparisons with other measuring instruments, they may evolve with more comprehensive models of the limiting factors in X-ray CT. • The measuring range and resolution limits of computed tomography (CT) are reviewed. • Experimental data is presented for determining measurement constraints of X-ray CT. • An X-ray CT instrument performance map is added to amplitude-wavelength (AW) space. • Several examples on the construction of AW maps for X-ray CT systems are provided. • AW maps facilitate comparisons of CT's competence with other measuring techniques. [ABSTRACT FROM AUTHOR]

• The digital twin concept is investigated using a discrete damped dynamic system model. • The concept of "slow time" is proposed to differentiate the evolution of the digital twin from its dynamics. • Stiffness and mass variation in the discrete system is considered. • Sensor errors are modeled in the update process of the digital twin. • The digital twin is expressed as a benchmark closed form solution. This paper employs a discrete damped dynamic system to investigate the emerging concept of a digital twin. Dynamic systems are well understood across engineering and science domains, and represent a familiar and convenient platform for exploring the various aspects of a digital twin design. The aim is to create a framework accessible to engineering sciences related to the aerospace, electrical, mechanical and computational area. The virtual model of the physical system is expressed as a differential equation in two-time scales, with the concept of a slow time being used to separate the evolution of the system properties from the instantaneous time. Cases involving stiffness variation and mass variation are considered, individually and together. It is assumed that the damped natural frequency and the time response are measured through sensors placed on the physical system. Issues of errors and reduced sampling rate in sensor measurements on the digital twin are investigated. The digital twin is expressed as an analytical solution through closed-form expressions and the effect of sensor errors is brought out through the simulations. Several key concepts introduced in this paper are summarized and ideas for urgent future research needs are proposed. The current work breaks free from the qualitative description of digital twins pervading the literature and can be used as benchmark solutions to validate digital twin of experimental dynamic systems and their implementation using sensors, the internet of things and deployment on the cloud computing systems. [ABSTRACT FROM AUTHOR]

In General-Purpose computing on Graphics Processing Unit (GPGPU), the use of CPUs is combined with that of GPUs. CPUs are used for sequential code, while GPUs are used for parallel code. GPGPU has been enabled by two key factors: (i) the massively parallel architecture of GPUs, which allows thousands of single cores to run parallel code; and (ii) the development of platforms, such as CUDA, that simplify implementing code for GPUs. GPGPU has established itself as the standard computing system in most computing fields due to the great improvements it brings. However, its use is not without problems, such as GPU underutilization, high cost, power consumption, etc. In this paper we present NGS (Network GPGPU System) to address the underutilization of GPUs in computing centers. NGS orchestrates the concurrent access to GPGPU resources from different nodes of the cluster by leveraging the remote GPU virtualization mechanism and the NVML library by NVIDIA. In this way, NGS enables different nodes of the cluster to access remote GPUs as if they were local at the same time that this access is guaranteed to be carried out without collisions. The main novelty is that NGS offers a global and standard solution independent of the computing environment used. Experimental results show up to 4x improvements compared to popular approaches. [ABSTRACT FROM AUTHOR]

This paper considers the dynamic response of a novel lightweight FOWT concept being developed by T-Omega Wind Ltd, that is able to float over even steep high waves, and be economical in deep water. The study aims to understand the response to waves during marine operations (installation, or maintenance) as part of optimizing its design. For this purpose real-time 6 degrees-of-freedom (6 DOF) simulations are computed for the system under operational and extreme sea wave scenarios in the state-of-the-art Multiphysics Marine Simulator at the National Decommissioning Centre (NDC). RAOs for heave and pitch displacements are evaluated across varying wave heights and periods of excitation to identify system behaviour including resonant frequencies. The model is calibrated by adjusting system damping parameters for each wave frequency to match experimental tests on a 1:60 scaled prototype at the Kelvin Hydrodynamics Laboratory, resulting in an ad hoc damping expression to produce appropriate system dynamic behaviour for "High" and "Low" Sea States. The study concludes by identifying ranges of wave parameters that limit peak motions, proposes analytical expressions for RAO responses and provides damping parameters that validate the Marine Simulator as a suitable tool to predict FOWT dynamic responses with reduced computation time. [Display omitted] • Dynamic response simulations of a novel FOWT concept under periodic sea wave excitation. • Virtual model of a novel low-mass FOWT concept developed for evaluation of marine operations. • Analytical RAO expressions for a novel FOWT with fitted hydrodynamic damping. • Wave parameter values providing better performance of a novel FOWT concept. • Frequency-dependent hydrodynamic damping of a novel FOWT to tune heave and pitch RAOs. [ABSTRACT FROM AUTHOR]

An accurate data-based prediction of the long-term evolution of Hamiltonian systems requires a network that preserves the appropriate structure under each time step. Every Hamiltonian system contains two essential ingredients: the Poisson bracket and the Hamiltonian. Hamiltonian systems with symmetries, whose paradigm examples are the Lie–Poisson systems, have been shown to describe a broad category of physical phenomena, from satellite motion to underwater vehicles, fluids, geophysical applications, complex fluids, and plasma physics. The Poisson bracket in these systems comes from the symmetries, while the Hamiltonian comes from the underlying physics. We view the symmetry of the system as primary, hence the Lie–Poisson bracket is known exactly, whereas the Hamiltonian is regarded as coming from physics and is considered not known, or known approximately. Using this approach, we develop a network based on transformations that exactly preserve the Poisson bracket and the special functions of the Lie–Poisson systems (Casimirs) to machine precision. We present two flavors of such systems: one, where the parameters of transformations are computed from data using a dense neural network (LPNets), and another, where the composition of transformations is used as building blocks (G-LPNets). We also show how to adapt these methods to a larger class of Poisson brackets. We apply the resulting methods to several examples, such as rigid body (satellite) motion, underwater vehicles, a particle in a magnetic field, and others. The methods developed in this paper are important for the construction of accurate data-based methods for simulating the long-term dynamics of physical systems. [ABSTRACT FROM AUTHOR]

Protecting data privacy is essential in developing practical cloud-based control systems to benefit from remote and distributed computation while avoiding the risks of disclosing sensitive information to potentially untrusted parties. This paper proposes a privacy-preserving framework to outsource the computation of model predictive control law to several untrusted cloud-based servers. A secret sharing scheme is employed to maintain the privacy of the system's data in all the stages of the control loop in a secure multi-party computation environment. In this regard, a privacy-preserving algorithm is derived to securely implement a projected gradient method to solve the underlying optimization problem, without revealing private data to external eavesdroppers and curious cloud-based servers. The proposed method is solely based on secret sharing, and all computations can be performed securely by cloud servers without the need to designate target nodes or engage the system's actuator in computing the intermediate steps. Therefore, the proposed method is less computationally demanding than the existing results based on homomorphic encryption and is applicable to systems with limited computing resources. Information-theoretic privacy assessments based on mutual information measures are provided. The efficacy of the proposed method is investigated on the cruise control problem of a freight train system. [Display omitted] • A secure multiparty computation framework is proposed to obtain a privacy-preserving model predictive control law. • The sensitive data is protected by employing a real-number secret sharing scheme, avoiding quantization errors. • All the computations are carried out securely at the cloud level. • The proposed method requires smaller message space and less computational load than the existing methods. [ABSTRACT FROM AUTHOR]

Convolution is a fundamental and computationally expensive primitive and finds ubiquitous in deep neural networks (DNNs). The evolving DNNs have spurred the emergence of numerous accelerators and they successfully achieve high throughput. However, for DNN inference with small batch sizes, the computational resources of the accelerators are often under-utilized, and the overhead of offloading is significant. Compared to accelerators, the CPU can better meet fast response requirements of inference, flexibly handle various models, and is suitable for various scenarios (from edge to data center). Therefore, CPU remains an attractive platform for DNN inference, despite the sub-optimal performance, and resource efficiency. In this paper, we propose CUTE, a scalable CPU-centric and ultra-utilized tensor engine for convolutions. It co-designs data flow and hardware architecture to leverage the data reuse and parallelism of convolutions. CUTE is composed of several small tensor elements (TEs) and two-level buffers. It employs a decoupled access-execution architecture and greedy strategy to feed data to TEs, enabling it to achieve ultra utilization and great scalability. CUTE is tightly coupled with the CPU to minimize offloading latency, thereby providing efficient convolution computing capabilities for the system. Experimental results show that under the same bandwidth, CUTE achieves an average performance improvement of 3.8x compared with the CPU AVX512 unit and 1.6x compared with the CPU AMX unit. Besides, CUTE achieves a speedup of 7.0x and 3.9x over Nvidia V100 GPU and Eyeriss accelerator respectively, due to higher utilization of computing units. [ABSTRACT FROM AUTHOR]