Frequent Updates Research Articles

Distribution grid monitoring based on feature propagation using smart plugs

AbstractSmart home power hardware makes it possible to collect a large number of measurements from the distribution grid with low latency. However, the measurements are imprecise, and not every node is instrumented. Therefore, the measured data must be corrected and augmented with pseudo-measurements to obtain an accurate and complete picture of the distribution grid. Hence, we present and evaluate a novel method for distribution grid monitoring. This method uses smart plugs as measuring devices and a feature propagation algorithm to generate pseudo-measurements for each grid node. The feature propagation algorithm exploits the homophily of buses in the distribution grid and diffuses known voltage values throughout the grid. This novel approach to deriving pseudo-measurement values is evaluated using a simulation of SimBench benchmark grids and the IEEE 37 bus system. In comparison to the established GINN algorithm, the presented approach generates more accurate voltage pseudo-measurements with less computational effort. This enables frequent updates of the distribution grid monitoring with low latency whenever a measurement occurs.

Harnessing Disruptive Technologies: Strategic Approach to Retail Product Innovation

Innovations in the retail industry, like Amazon's Just Walk Out (JWO) technology, are changing the way people shop by providing a cashierless experience. This article examines how to strategically utilize disruptive technologies for retail product innovation, with JWO as a case study. The examination summarizes JWO technology, its uptake, and the obstacles encountered during implementation. The research also explores the business dynamics of technological advancement, underlining the crucial role of disruptive technologies in improving customer satisfaction, operational effectiveness, and product creativity. The document points out the main drawbacks of JWO technology, such as restricted scalability, expensive implementation and operational costs, and the requirement for frequent technology updates. Suggestions for enhancing JWO technology and its utilization include implementing a licensing or subscription-based business model, incorporating modular design, and enhancing flexibility. The research also covers tactical factors for merchants when adopting innovative technologies, highlighting the significance of prioritizing customer needs, evaluating costs and benefits, considering scalability, building partnerships, and incorporating other technologies. The paper ends by emphasizing the capability of disruptive technologies to change the retail sector and the requirement for companies to find a balance between innovation and practical factors to thrive in a competitive environment. Keywords—Disruptive Technologies, Retail Product Innovation, Just Walk Out (JWO), Customer Satisfaction, Operational Efficiency, Machine Learning, Artificial Intelligence(AI), Technology Strategy, Machine learning (ML), Data.

Teaching agile: A case study approach with an optional game

Learning Agile philosophy is essential for system analysis and design or software development courses. Organizations operate in highly dynamic environments, constantly needing to adapt, change, or create innovative Information Systems. Agile provides guiding values and principles that enable systems analysts to support organizations in responding to change and developing, modifying, or updating information systems. By learning Scrum, an Agile-inspired framework, students can assume the role of systems analyst, organizing activities and working in teams iteratively to solve complex problems using Agile principles. This hands-on approach helps students understand the importance of stakeholder involvement, constant scrutiny, and adjustment to achieve frequent system updates. Hence, this case study immerses students in a scenario where they apply the Scrum framework, with an optional puzzle game to introduce students to agile philosophy through its manifesto. Learning by doing has proven effective for complex or dry topics, with students responding positively to this method. Interview data collected at the end of the activity showed high student satisfaction and a strong sense of knowledge acquisition while enjoying the class. Additionally, integrating a game to internalize agile concepts with a teaching case to apply concepts presents a rich educational approach that can be adapted to other Information Systems topics.

ViralPrimer: a Web Server to Monitor Viral Nucleic Acid Amplification Tests' Primer Efficiency during Pandemics, with Emphasis on SARS-CoV-2 and Mpox.

Accurate pathogen identification is crucial during outbreaks, especially with the emergence of new variants requiring frequent primer updates. However, resources for maintaining up-to-date verification of primer sequences are often limited, which poses challenges for reliable diagnosis and hinders potential monitoring efforts based on genome sequencing. To address this, we introduce ViralPrimer, a web server facilitating primer design, SARS-CoV-2 and Mpox variant monitoring, and adaptation to future threats. ViralPrimer aims to enhance diagnostic accuracy with its comprehensive primer database, mutation analysis assistance, and user primer upload feature. Its adaptable design allows monitoring of other rapidly mutating pathogens, contributing to broader public health protection efforts. ViralPrimer is freely accessible and open to all users with no login requirement at https://viralprimer.elte.hu/. The application is hosted on a DJANGO v3.2.13 web server, with a PostgreSQL database, and the frontend was implemented using jQuery v3.6.0, vanilla JavaScript vES6, and Bootstrap v5.1. https://viralprimer.elte.hu/help#downloads.

Event-triggered iterative learning containment control for one-sided Lipschitz nonlinear singular switched multi-agent systems with multiple leaders and error quantization

In this work, under a repeatable control environment, the iterative learning containment control issue for multi-agent systems with one-sided Lipschitz nonlinear singular switched dynamics is addressed. It focuses on the output containment for one-sided Lipschitz nonlinear singular switched multi-agent systems (SWMASs) with multiple leaders, for which an event-triggered containment quantized iterative learning control (QILC) scheme is designed. Due to the limited system resources, SWMASs are often restricted in their communication capabilities and the frequency of controller updates in many applications, the proposed event-triggering mechanism can greatly restrict the use of communication and computation resources throughout the iterative learning process. When the triggering requirement is met, the event-triggered QILC input is repeatedly updated; alternatively, the control input stays unchanged. To conquer the nonidentical initial states of followers, an initial state learning with error quantization and event-triggering is also utilized. By virtue of some appropriate learning gains and sufficient convergent conditions, the containment error convergence can be accomplished for one-sided Lipschitz SWMASs. Finally, some numerical examples via Matlab are provided to demonstrate the efficacy of the event-triggered QILC scheme, it indicates that followers’ outputs can converge to the convex hull formed by leaders and the containment errors gradually converge to the 10−3 - vicinity of the origin along the iteration axis, where the communication resources can save more than 50% in the transmission channel.

Heuristic dense reward shaping for learning-based map-free navigation of industrial automatic mobile robots

This paper presents a map-free navigation approach for industrial automatic mobile robots (AMRs), designed to ensure computational efficiency, cost-effectiveness, and adaptability. Utilizing deep reinforcement learning (DRL), the system enables real-time decision-making without fixed markers or frequent map updates. The central contribution is the Heuristic Dense Reward Shaping (HDRS), inspired by potential field methods, which integrates domain knowledge to improve learning efficiency and minimize suboptimal actions. To address the simulation-to-reality gap, data augmentation with controlled sensor noise is applied during training, ensuring robustness and generalization for real-world deployment without fine-tuning. Training results underscore HDRS’s superior convergence speed, training stability, and policy learning efficiency compared to baselines. Simulation and real-world evaluations establish HDRS-DRL as a competitive alternative, outperforming traditional approaches, and offering practical applicability in industrial settings.

Challenges and guidance for implementing social distancing for COVID-19 in care homes: a mixed methods rapid review.

Older people living in care homes are at high risk of poor health outcomes and mortality if they contract coronavirus disease 2019. Protective measures include social distancing and isolation, although implementation is challenging. To explore the real-life experiences of social distancing and isolation in care homes for older people, and to develop a toolkit of guidance and resources. A mixed-methods, phased design. Six care homes in England caring for older adults. Care home staff (n = 31), residents (n = 17), family members (n = 17), senior health and care leaders (n = 13). A rapid review to assess the social distancing and isolation measures used by care homes to control the transmission of coronavirus disease 2019 and other infectious diseases (phase 1), in-depth case studies of six care homes, involving remote individual interviews with staff, residents and families, collection of policies, protocols and routinely collected care home data, remote focus groups with senior health and care leaders (phase 2) and stakeholder workshops to co-design the toolkit (phase 3). Interview and focus group data and care home documents were analysed using thematic analysis and care home data using descriptive statistics. The rapid review of 103 records demonstrated limited empirical evidence and the limited nature of policy documentation around social distancing and isolation measures in care homes. The case studies found that social distancing and isolation measures presented moral dilemmas for staff and often were difficult, and sometimes impossible to implement. Social distancing and isolation measures made care homes feel like an institution and denied residents, staff and families of physical touch and other forms of non-verbal communication. This was particularly important for residents with cognitive impairment. Care homes developed new visiting modalities to work around social distancing measures. Residents and families valued the work of care homes to keep residents safe and support remote communication. Social distancing, isolation and related restrictions negatively impacted on residents' physical, psychological, social and cognitive well-being. There were feelings of powerlessness for families whose loved ones had moved into the care home during the pandemic. It was challenging for care homes to capture frequent updates in policy and guidance. Senior health and care leaders shared that the care home sector felt isolated from the National Health Service, communication from government was described as chaotic, and trauma was inflicted on care home staff, residents, families and friends. These multiple data sources have informed the co-design of a toolkit to care for residents, families, friends and care home staff. The review included papers published in English language only. The six care homes had a Care Quality Commission rating of either 'good' or 'outstanding'. There was a lack of ethnic diversity in resident and family participants. Care homes implemented innovative approaches to social distancing and isolation with varying degrees of success. A legacy of learning can help rebuild trust at multiple levels and address trauma-informed care for residents, families, friends and staff. Future work can include evaluation of the toolkit, research to develop a trauma-informed approach to caring for the care home sector and co-designing and evaluating an intervention to enable residents with different needs to transition to living well in a care home. This award was funded by the National Institute for Health and Care Research (NIHR) Health and Social Care Delivery Research programme (NIHR award ref: NIHR132541) and is published in full in Health and Social Care Delivery Research; Vol. 12, No. 45. See the NIHR Funding and Awards website for further award information.

Assessment of knowledge and perception of prescribers towards rational medicine use in the Ashanti Region of Ghana.

Prescribers must possess extensive knowledge and maintain a positive attitude towards the rational use of medicines to achieve desirable treatment outcomes and effectively prevent treatment failures, increased costs, drug toxicities, and interactions. The objective of this study was to evaluate prescribers' understanding and perception concerning the rational use of medicines in public hospitals. Additionally, the study aimed to identify the factors that influence rational prescribing practices. A structured data instrument was developed to collect demographic data and evaluate participants' knowledge and perception of rational medicine use, in line with the study objectives. Chi-squared statistics and Fisher's exact test were utilized to identify factors associated with good knowledge and perception among participants. Logistic regression was then employed to assess the strength of the associations, with odd ratios reported at a significant level of 0.05. Out of 192 participants, 85.4% held a positive view of rational medicine use, stressing patient safety and recognizing risks like antimicrobial resistance and polypharmacy. Perception was influenced by factors such as prescriber profession, access to references, and drug bulletin updates. Additionally, 65.6% demonstrated good knowledge of rational medicine use, which was notably influenced by factors like using standard prescribing guidelines, having a functional Drug and Therapeutics Committee, prescriber profession, and the frequency of drug bulletin updates. The study emphasizes the critical need to address knowledge gaps among healthcare professionals, especially nurses and other prescribers, to ensure the safe and effective use of medications. It highlights the positive influence of utilizing preferred prescribing references and the existence of functional Drug and Therapeutics Committees in hospitals on knowledge levels. However, the unexpected findings regarding the limited impact of frequent updates of drug bulletins require further investigation.

A Deep Learning Approach for Enhanced Real-Time Prediction of Winter Road Surface Temperatures in High-Altitude Mountain Areas

Low temperatures and icing in winter are significant factors that severely affect highway safety and traffic mobility. To enhance the precision and reliability of real-time winter road surface temperature (RST) prediction, a short-term prediction model is developed that harnesses both feature selection and deep learning. Leveraging meteorological data from a mountain highway in Yunnan, China, the key environmental variables affecting road surface temperature were first extracted using a random forest (RF) model for feature selection. These features were then combined with RST data to construct multiple groups of input variable combinations for the prediction model. A short-term prediction model with a 10-minute update frequency was built using a long short-term memory neural network (LSTM), namely RF-LSTM. The best input variable combination and preset parameters for the prediction model were determined through comparative testing with prevalent machine learning models, and the transferability of the prediction model was verified. The results showed that the best input variable combination for the RF-LSTM prediction model was road surface temperature and air temperature. The model recognised that the short-term RST was affected by long and short-term memory characteristics within a two-hour timeframe. When compared to the RF model, backpropagation (BP) neural network model and the standard LSTM model, the proposed model reduces prediction errors by 59.15%, 31.10% and 20.26%, respectively, while the prediction accuracy is 99.13% within an error margin of ±0.5℃. On the verification dataset, the proposed model maintains its time transferability with an average prediction absolute error of 0.0478. In all, the proposed model not only achieves a higher level of precision in real-time RST predictions but also ensures a more consistent and reliable performance under the challenging conditions of high altitude and mountainous terrain, offering enhanced support for traffic safety and road maintenance decision-making.

Overcoming Product Catalog Challenges in Telecom : A Technical Perspective

This comprehensive article explores the intricate challenges telecom providers face in managing complex product catalogs across multiple customer touchpoints. With the global telecom market projected to reach $2.47 trillion by 2028, efficient catalog management is crucial for maintaining competitiveness and customer satisfaction. The article delves into the multifaceted complexities of telecom product catalogs, including regional pricing variations, dynamic promotional offers, and the need for real-time synchronization across platforms. It outlines key challenges such as catalog complexity, slow data retrieval, frequent updates, third-party integration latencies, high traffic spikes, and campaign management inconsistencies. The article then presents innovative technical solutions and best practices, including the adoption of industry-leading catalog solutions, leveraging ElasticSearch for faster data retrieval, implementing differential processing for updates, preprocessing third-party integrations, utilizing dynamic UI templates, employing advanced campaign management tools, and implementing inventory prefetching. These solutions aim to enhance operational efficiency, improve customer experiences, and accelerate time-to-market in the rapidly evolving telecom industry.

Utilizing graph neural networks for adverse health detection and personalized decision making in sensor-based remote monitoring for dementia care

Background:Sensor-based remote health monitoring is increasingly used to detect adverse health in people living with dementia (PLwD) at home, aiming to prevent hospitalizations and reduce caregiver burden. However, home sensor data is often noisy, overly granular, and suffers from unreliable labeling, data drift and high variability between households. Current anomaly detection methods lack generalizability and personalization, often requiring anomaly-free training data and frequent model updates. Objective:To develop a lightweight, explainable, self-supervised approach with personalized alert thresholds to detect adverse health events in PLwD, using changes in home activity. Methods:We hypothesized that health downturns manifest as detectable shifts in household movement patterns. Our approach leverages a Graph Barlow Twins contrastive model, which uses granular activity data and a macroscopic view to extract noise-robust, high-level and low-level discriminative features that represent daily activity patterns. Household-personalized alert thresholds are calculated based on clinician-set target alert rates, and daily anomaly scores are compared against these thresholds, triggering alerts for the clinical monitoring team. Model attention weights support explainability. Data were collected from a real-world dataset by the UK Dementia Research Institute (August 2019-April 2022). Results:Our model outperformed state-of-the-art temporal graph algorithms in detecting agitation and fall events across three patient cohorts, achieving 81% average recall and 88% generalizability at a target alert rate of 7%. Conclusion:We developed a novel, lightweight, explainable, and personalized Graph Barlow Twins model for real-world remote health monitoring in dementia care, with potential for broader applications in healthcare and sensor-based environments.

An optimal parameterized Newton-type structure-preserving doubling algorithm for impact angle guidance-based 3D pursuer/target interception engagement

The proposed strategy, finite-time state-dependent Riccati equation (FT-SDRE)-based impact angle guidance, is generally employed to solve the 3D pursuer/target interception model with fixed lateral accelerations. This article expands its application to a general scenario where the lateral acceleration of a target may change. To achieve this, we approximate the accelerations of the azimuth and elevation angles of the target in the inertial frame via second-order finite difference schemes and develop a high-performance FT-SDRE algorithm with structure-preserving doubling algorithms (SDAs). As a result, the update frequency of the controller can be increased, and better guidance of the pursuer can be obtained to address the high maneuverability of the target during the entire interception procedure. At every state of the FT-SDRE, a modified Newton–Lyapunov method is employed to solve the continuous algebraic Riccati equation (CARE), and a new simplified SDA with adaptive optimal parameter selection is proposed for solving the associated Lyapunov equation. Our numerical results demonstrate that the FT-SDRE algorithm accelerated by our proposed methods is approximately three times faster than the FT-SDRE algorithm, in which the MATLAB functions icare and lyap are used to solve the CARE and the Lyapunov equation, respectively, throughout the entire interception procedure. In other words, the control frequency can be increased threefold. In our benchmark cases where the target maneuvers with nonlinear lateral acceleration, the target can be intercepted earlier via the proposed FT-SDRE algorithm.

High resolution (1-km) surface soil moisture generation from SMAP SSM by considering its difference between freezing and thawing periods in the source region of the Yellow River

Soil moisture (SM) is a critical component of the land surface hydrological cycle, significantly impacting various sectors such as hydrology, meteorology, and agriculture. Accurate, high-resolution SM data are essential for effective flood forecasting, water resource management, and understanding the soil freeze-thaw processes in cold regions. This study aims to generate 1 km resolution liquid surface SM (SSM) data with a twice-daily update frequency by downscaling SMAP Level-4 SSM data using random forest (RF) and multiple linear regression (MLR) in the source region of the Yellow River (SRYR), by considering the differences in SM changes between freezing and thawing periods. To obtain the SSM data, 16 downscaling schemes of both RF and MLR were designed for each of the three scenarios. In each downscaling process, both land surface temperature (LST) and normalized difference vegetation index (NDVI) were utilized in MLR and RF models, alongside various combinations of additional variables such as albedo, elevation, leaf area index (LAI), soil texture. Results showed that during the freezing period, RF produced superior SSM estimates when supplemented with NDVI, LST, albedo, elevation, LAI, and soil texture. MLR was more effective during the thawing period when paired with NDVI, LST, elevation, LAI, and soil texture. During the freezing period, the downscaled SMAP SSM exhibited average R, RMSE, ubRMSE of 0.76, 0.029 m3·m-3, and 0.023 m3·m-3, respectively, when compared with in-situ measurements. During the thawing period, the average R, MAE, RMSE, and ubRMSE between the downscaled SMAP SSM and in-situ measurements were 0.52, 0.057 m3·m-3, 0.067 m3·m-3, and 0.054 m3·m-3, respectively, compared to 0.45, 0.070 m3·m-3, 0.083 m3·m-3, and 0.060 m3·m-3 for the original SMAP SSM. Thus, the research significantly enhances both the accuracy and spatial resolution of SMAP SSM estimations, underscoring its vital role in advancing hydrological studies within the SRYR.

Aerosol immunization with influenza matrix, nucleoprotein, or both prevents lung disease in pig

Current influenza vaccines are strain-specific and require frequent updates to combat new strains, making a broadly protective influenza vaccine (BPIV) highly desirable. A promising strategy is to induce T-cell responses against internal proteins conserved across influenza strains. In this study, pH1N1 pre-exposed pigs were immunized by aerosol using viral vectored vaccines (ChAdOx2 and MVA) expressing matrix (M1) and nucleoprotein (NP). Following H3N2 challenge, all immunizations (M1, NP or NPM1) reduced lung pathology, but M1 alone offered the greatest protection. NP or NPM1 immunization induced both T-cell and antibody responses. M1 immunization generated no detectable antibodies but elicited M1-specific T-cell responses, suggesting T cell-mediated protection. Additionally, a single aerosol immunization with the ChAdOx vaccine encoding M1, NP and neuraminidase reduced lung pathology. These findings provide insights into BPIV development using a relevant large natural host, the pig.

Sustainable Supply Chain Development Strategies in the Fast Fashion Industry: A Case Study of Nike

With the acceleration of globalization, companies are increasingly concerned about environmental protection and social responsibility, and sustainable development has become an important trend in industrial development. The fast fashion industry has attracted much attention due to its rapid market response, mass production, and frequent style updates. However, its supply chain faces many challenges in terms of sustainability. Based on this background, this study delves into sustainable supply chain management in the fast fashion industry using Nike as a case study. By analyzing Nikes strategies, practices and achievements, this paper provides useful suggestions and insights for achieving sustainability in the fast fashion industry. Nike has positively contributed to supply chain sustainability by promoting a circular economy, establishing a green supply chain and optimizing inventory management. This study can provide an important reference for the sustainable development of the fast fashion industry and a guide for companies to innovate and improve continuously under the global sustainability agenda.

Hepatitis B infection: Evaluation of demographics and treatment of chronic hepatitis B infection in Northern-western Tanzania.

Chronic hepatitis B virus (HBV) infection is still a major public health problem. In response to the World Health Organization (WHO), Tanzania implemented immunization and treatment to achieve the eradication of HBV infection by 2030. To achieve this goal, frequent updates of demographic data, antiviral therapy eligibility, and uptake are essential. We therefore evaluated demographic data, antiviral therapy eligibility, and uptake among chronically HBV-infected patients attending at Bugando Medical Centre (BMC), Tanzania. A cross-sectional study enrolled 196 chronic HBV patients from April 23, 2023, to October 10, 2023, at BMC, where 100 and 96 patients were retrospectively and prospectively enrolled, respectively. Study's ethical clearance and permission were observed by the Catholic University of Health and Allied Sciences/Bugando Medical Centre research ethics and review committee and the Bugando Medical Centre management respectively. For all patients, socio-demographic data and whole blood samples were obtained. Full blood picture, alanine and aspartate amino transferases, and HBV viral load parameters were determined. Aspartate-Platelet Ratio Index (APRI) and Fibrosis Four (FIB-4) scores were calculated according to their respective formulas. Therapy eligibility and uptake were evaluated according to the 2015 WHO HBV prevention, treatment, and care guidelines. The data were summarized and analysed using STATA version 15. The median age for all patients was 39 [IQR: 32-47.5] years. Nearly all study patients, 99% (194/196), were older than 20 years old, with significant male dominance (73.5% [144/196] versus 26.5% [52/196]; p<0.0001). Anti-HBV antiviral therapy eligibility was 22.4%, while uptake was 6.8% (3/4), which was significantly lower than the WHO expectation of 80% (p <0.0001). Almost all chronically HBV-infected patients attending at BMC were older than 20 years old and were significantly dominated by males. Antiviral therapy uptake was remarkably lower than expected by the WHO towards combating HBV infection by 2030.

PPP time transfer using an adaptive clock constraint model

ABSTRACT In the processing of Precise Point Positioning (PPP) data, the receiver clock is approached with nearly infinite uncertainty, rendering it difficult to fulfill the requirements of high-precision time frequency applications. Therefore, a receiver clock model is essential. In this study, we first analyze the “over-constraint” problem in the existing clock model and subsequently propose a new clock model, called the Adaptive Clock Constraint (ACC) model, which relies on a sliding window to update covariance and frequency characteristics parameters in real-time. To verify the robustness of the ACC model, three experiments were conducted, and the results show that the ACC model not only is suitable for different types of atomic clock stations but also has superior frequency stability and time transfer precision in contrast to the BIPM PPP, the IGS products and White Noise (WN) model results. Using the optical fiber results as a time reference, the STD of the time difference between the ACC model and optical fiber results is 0.13 ns and the frequency stability is 1.28 × 10−16 on average for one week, representing improvements of more than 10% and 15% compared with the BIPM PPP results.

Dynamic event-triggered finite-horizon robust suboptimal control of multi-player systems with input disturbances

This paper presents an adaptive dynamic event-triggered robust finite-horizon suboptimal control scheme for multi-player nonzero-sum game of nonlinear systems with input disturbances based on dynamic programming (ADP) and integral sliding mode (ISM) control techniques. Firstly, dynamic event-triggered adaptive ISM controllers are designed to handle input disturbances that can force the system state to remain on the sliding manifold and relax the known upper bounds condition of input disturbances. Corresponding event-triggered condition is employed for ISM controller of each player to reduce the update frequency. Then, utilizing the obtained ISM dynamics, we employ ADP-based single critic neural networks to approximate the time-varying solution of the Hamilton–Jacobi equations. Thus, the event-triggered finite-horizon suboptimal controllers can be obtained, and corresponding state-based dynamic event-triggered condition is designed to improve the resource utilization. The stability property of the closed-loop system is demonstrated through the Lyapunov technique. Finally, the effectiveness of the proposed control scheme is confirmed through two simulation examples.

Continual learning with high-order experience replay for dynamic network embedding

Dynamic network embedding (DNE) poses a tough challenge in graph representation learning, especially when confronted with the frequent updates of streaming data. Conventional DNEs primarily resort to parameter updating but perform inadequately on historical networks, resulting in the problem of catastrophic forgetting. To tackle such issues, recent advancements in graph neural networks (GNNs) have explored matrix factorization techniques. However, these approaches encounter difficulties in preserving the global patterns of incremental data. In this paper, we propose CLDNE, a Continual Learning framework specifically designed for Dynamic Network Embedding. At the core of CLDNE lies a streaming graph auto-encoder that effectively captures both global and local patterns of the input graph. To further overcome catastrophic forgetting, CLDNE is equipped with an experience replay buffer and a knowledge distillation module, which preserve high-order historical topology and static historical patterns. We conduct experiments on four dynamic networks using link prediction and node classification tasks to evaluate the effectiveness of CLDNE. The outcomes demonstrate that CLDNE successfully mitigates the catastrophic forgetting problem and reduces training time by 80% without a significant loss in learning new patterns.

Non-fragile event-triggered control for switched positive linear systems using state-dependent switching method

This article is concerned with the issues of stability and L1-gain characteristic for switched positive linear systems utilizing bumpless transfer (BT) technique and event-triggered (ET) strategy. Under the state-driven switching policy, a regulatory approach that couples BT and ET is introduced to attenuate the hybrid bumps stemming from switching and triggering moments. Compared with traditional state-driven switching, the proposed hysteresis switching can delay the occurrence of switching. Based on multiple copositive Lyapunov functions, a state feedback controller with an ET mechanism is constructed for positive systems, which reduces the update frequency of the actuator, and this control strategy is extended to the case of actuator faults. It is confirmed that the proposed control scheme can assure the positivity, stability, and L1-gain performance of the system. Additionally, the Zeno phenomenon is addressed via establishing a positive lower bound for the ET interval. Ultimately, simulation examples containing a turbofan engine model are provided to demonstrate the effectiveness of the proposed approach.