Differential equations are involved in modeling many engineering problems. Many efforts have been devoted to solving differential equations. Due to the flexibility of neural networks, Physics Informed Neural Networks (PINNs) have recently been proposed to solve complex differential equations and have demonstrated superior performance in many applications. While the L2 loss function is usually a default choice in PINNs, it has been shown that the corresponding numerical solution is incorrect and unstable for some complex equations. In this work, we propose a new PINNs framework named Kernel Packet accelerated PINNs (KP-PINNs), which gives a new expression of the loss function using the reproducing kernel Hilbert space (RKHS) norm and uses the Kernel Packet (KP) method to accelerate the computation. Theoretical results show that KP-PINNs can be stable across various differential equations. Numerical experiments illustrate that KP-PINNs can solve differential equations effectively and efficiently. This framework provides a promising direction for improving the stability and accuracy of PINNs-based solvers in scientific computing.

Artificial intelligence (AI) has immense potential in time series prediction, but most explainable tools have limited capabilities in providing a systematic understanding of important features over time. These tools typically rely on evaluating a single time point, overlook the time ordering of inputs, and neglect the time-sensitive nature of time series applications. These factors make it difficult for users, particularly those without domain knowledge, to comprehend AI model decisions and obtain meaningful explanations. We propose CGS-Mask, a post-hoc and model-agnostic cellular genetic strip mask-based saliency approach to address these challenges. CGS-Mask uses consecutive time steps as a cohesive entity to evaluate the impact of features on the final prediction, providing binary and sustained feature importance scores over time. Our algorithm optimizes the mask population iteratively to obtain the optimal mask in a reasonable time. We evaluated CGS-Mask on synthetic and real-world datasets, and it outperformed state-of-the-art methods in elucidating the importance of features over time. According to our pilot user study via a questionnaire survey, CGS-Mask is the most effective approach in presenting easily understandable time series prediction results, enabling users to comprehend the decision-making process of AI models with ease.

Negative emotional states, such as anxiety and depression, pose significant challenges in contemporary society, often stemming from the stress encountered in daily activities. Stress (state or level) recognition is a crucial prerequisite for effective stress management and intervention. Presently, wearable devices have been employed to capture physiological signals and analyze stress states. However, their constant skin contact can lead to discomfort and disturbance during prolonged monitoring. In this paper, a peak attention-based multitasking framework is presented for non-contact stress recognition. The framework extracts rPPG signals from RGB facial videos, utilizing them as inputs for a novel multi-task attentional convolutional neural network for stress recognition (MTASR). It incorporates peak detection and HR estimation as auxiliary tasks to facilitate stress recognition. By leveraging multi-task learning, MTASR can utilize information related to stress physiological responses, thereby enhancing feature extraction efficiency. For stress recognition, two binary classification tasks are applied: stress state recognition and stress level recognition. The model is validated on the UBFC-Phys public dataset and demonstrates an accuracy of 94.33% for stress state recognition and 83.83% for stress level recognition. The proposed method outperforms the dataset's baseline methods and other competing approaches.

Intraoperative hypotension (IOH) events warning plays a crucial role in preventing postoperative complications, such as postoperative delirium and mortality. Despite significant efforts, two fundamental problems limit its wide clinical use. The well-established IOH event warning systems are often built on proprietary medical devices that may not be available in all hospitals. The warnings are also triggered mainly through a predefined IOH event that might not be suitable for all patients. This work proposes a composite multi-attention (CMA) framework to tackle these problems by conducting short-term predictions on user-definable IOH events using vital signals in a low sampling rate with demographic characteristics. Our framework leverages a multi-modal fusion network to make four vital signals and three demographic characteristics as input modalities. For each modality, a multi-attention mechanism is used for feature extraction for better model training. Experiments on two large-scale real-world data sets show that our method can achieve up to 94.1% accuracy on IOH events early warning while the signals sampling rate is reduced by 3000 times. Our proposal CMA can achieve a mean absolute error of 4.50 mm Hg in the most challenging 15-minute mean arterial pressure prediction task and the error reduction by 42.9% compared to existing solutions.

Machine learning models are increasingly used in time series prediction with promising results. The model explanation of time series prediction falls behind the model development and makes less sense to users in understanding model decisions. This paper proposes ES-Mask, a post-hoc and model-agnostic evolutionary strip mask-based saliency approach for time series applications. ES-Mask designs the mask consisting of strips with the same salient value in consecutive time steps to produce binary and sustained feature importance scores over time for easy understanding and interpretation of time series. ES-Mask uses an evolutionary algorithm to search for the optimal mask by manipulating strips in rounds, thus is agnostic to models by involving no internal model states in the search. The initial experiments on MIMIC-III data set show that ES-Mask outperforms state-of-the-art methods.

This article investigates the nonsynchronous-observer -based output-feedback control of nonlinear systems by Takagi–Sugeno fuzzy affine models. In order to estimate the unmeasurable system states and handle the nonsynchronization issue of premise variables between the observer/controller and the original system, a new nonsynchronous piecewise fuzzy affine observer is first synthesized to estimate the unmeasurable system state variables. Via utilizing some elegant matrix inequalities with convexification techniques, new observer-based piecewise output-feedback controller design results are proposed within the framework of piecewise quadratic Lyapunov functions. Simulation studies are given to illustrate the validity of the developed approach.

In this study, we consider a multiway massive multi-input multi-output (MIMO) relay network over Rician fading channels, where all users intend to share their information with the other users via amplify-and-forward (AF) relays equipped with a great number of antennas. More practical, the imperfect channel state information (CSI) is taken into account. To evaluate the performance of the considered networks, we derived an analytical approximation expression for the spectral efficiency with zero-forcing (ZF) receivers in a closed form. To obtain more insights, the asymptotic analysis as the number of relay antenna approaching infinity is carried out. Finally, the power scaling law is analyzed for two scenarios. The results reveal that (1) massive MIMO is capable of compensating the loss caused by Rician fading, (2) the sum spectral efficiency increases with the increase of the Rician factor, and (3) deploying large-scale antenna is effective to save cost and keep performance.

The finite-time coverage control problem for networked mobile agents with continuous-time dynamics and unidirectional motion constraint on a closed curve is addressed in this article. The objective is to minimize a coverage cost function in finite time which characterizes the largest arrival time from the multiagent system to any point on the curve. Low gain feedback is employed to design the distributed coverage control laws for the agents with different input constraints due to their limited movement capabilities. An upper bound on each low gain is also provided by showing that the distance between the neighboring agents is always lower and upper bounded. Under the proposed control laws, the networked mobile agents can be driven to the optimal configuration, minimizing the coverage cost function in finite time. Due to the unidirectional motion of the agents, there exists at least one agent which remains static throughout the coverage task. As a result, a less conservative upper bound on the low gains is derived, which provides the possibility of better convergence rates of the proposed coverage control laws.

This work studies the event-triggered control problem for networked control systems. The plant is controlled directly by a dynamic local controller, which receives the reference control signal from the remote controller. The measurement signal of the plant and the reference control signal of the remote controller has their separate event trigger, and thus the remote controller and the local controller can decide when to transmit signals on their own. It is proved that with the proposed control approach and the dual event trigger, the closed loop system is globally asymptotically stable, which has been illustrated by simulation results.

Cloud manufacturing (CMfg) has become a brand-new service-oriented networked manufacturing. The key technology that achieves this manufacture mode is to aggregate all kinds of potential CMfg services to create value-added composite CMfg services, so as to satisfy various manufacturing requirements. As massive CMfg services with the same manufacturing functionalities and different quality of service (QoS) are emerging on the Internet, CMfg services dynamic composition becomes more and more complex. One challenge of this technology is how to efficiently solve the problem of multipath CMfg services dynamic composition. Another challenge is how to conquer the overconstrained problem occurred in the process of CMfg services dynamic composition. To tackle these two issues, this paper proposes an approach for multipath CMfg services dynamic composition based on improved QoS constraint hierarchies (CHs) model. First, a QoS constraints description model with new evaluation function is propounded based on the CHs theory to conquer the overconstrained problem; second, improved max–min ant system (MMAS) is put forward by introducing improved transfer rule, new global pheromone updating rule and mutation mechanism into MMAS, and then an optimization algorithm named culture genetic MMAS is constructed for multipath CMfg services dynamic composition. Finally, experimental results and performance comparison prove that the proposed method is both effective and efficient.

article Free Access Share on Optimizing bulk data transfer performance: a packet train model Authors: C. Song Univ. of Wisconsin, Madison Univ. of Wisconsin, MadisonView Profile , L. H. Landweber Univ. of Wisconsin, Madison Univ. of Wisconsin, MadisonView Profile Authors Info & Claims ACM SIGCOMM Computer Communication ReviewVolume 18Issue 4August 1988 pp 134–145https://doi.org/10.1145/52325.52339Published:01 August 1988Publication History 8citation670DownloadsMetricsTotal Citations8Total Downloads670Last 12 Months38Last 6 weeks3 Get Citation AlertsNew Citation Alert added!This alert has been successfully added and will be sent to:You will be notified whenever a record that you have chosen has been cited.To manage your alert preferences, click on the button below.Manage my AlertsNew Citation Alert!Please log in to your account Save to BinderSave to BinderCreate a New BinderNameCancelCreateExport CitationPublisher SiteeReaderPDF