Dynamic Task Research Articles

Dynamic task offloading and service caching based on game theory in vehicular edge computing networks

Vehicular edge computing (VEC) enables task offloading from vehicle to edge servers deployed on Road Side Units (RSUs) and to vehicle with idle resources, which is considered as a promising vehicle network architecture. However, efficient task offloading and task resource cache management in the VEC network is challenging. In this work, we formulate a joint task offloading and service caching problem (JTOSCP) for VEC with edge-vehicle cooperation, aiming to minimize the total task delay-energy value and cost value of all vehicles while guaranteeing task processing delay tolerance. The formulated JTOSCP is proven as NP-hard problem. Due to the complexity of JTOSCP, we propose a cooperative task service caching and task offloading algorithm named CO-MATCH. In this algorithm, we design the dynamic programming-based service caching (DPSC) Algorithm to incentivize edge services and volunteer vehicles to cache, which takes into account both service caching preferences and social similarity. The Many-to-One Matching Game (MOMG) Algorithm is proposed to stimulate edge device games and achieve the optimal task offloading. Then, the CO-MATCH method is proven to be stable. Simulation results demonstrate that our proposed approach can significantly improve the quality of service of user vehicles compared to the other methods.

Full-body kinematics and head stabilisation strategies during walking in patients with chronic unilateral and bilateral vestibulopathy

Chronic imbalance is a frequent and limiting symptom of patients with chronic unilateral and bilateral vestibulopathy. A full-body kinematic analysis of the movement of patients with vestibulopathy would provide a better understanding of the impact of the pathology on dynamic tasks such as walking. Therefore, this study aimed to investigate the global body movement during walking, its variability (assessed with the GaitSD), and the strategies to stabilise the head (assessed with the head Anchoring Index). The full-body motion capture data of 10 patients with bilateral vestibulopathy (BV), 10 patients with unilateral vestibulopathy (UV), and 10 healthy subjects (HS) walking at several speeds (slow, comfortable, and fast) were analysed in this prospective cohort study. We observed only a few significant differences between groups in parts of the gait cycle (shoulder abduction–adduction, pelvis rotation, and hip flexion–extension) during the analysis of kinematic curves. Only BV patients had significantly higher gait variability (GaitSD) for all three walking speeds. Head stabilisation strategies depended on the plan of motion and walking speed condition, but BV and UV patients tended to stabilise their head in relation to the trunk and HS tended to stabilise their head in space. These results suggest that GaitSD could be a relevant biomarker of chronic instability in BV and that the head Anchoring Index tends to confirm clinical observations of abnormal head-trunk dynamics in patients with vestibulopathy while walking.

RN‐Net: Reservoir Nodes‐Enabled Neuromorphic Vision Sensing Network

Neuromorphic computing systems promise high energy efficiency and low latency. In particular, when integrated with neuromorphic sensors, they can be used to produce intelligent systems for a broad range of applications. An event‐based camera is such a neuromorphic sensor, inspired by the sparse and asynchronous spike representation of the biological visual system. However, processing the event data requires either using expensive feature descriptors to transform spikes into frames, or using spiking neural networks (SNNs) that are expensive to train. In this work, a neural network architecture is proposed, reservoir nodes‐enabled neuromorphic vision sensing network (RN‐Net), based on dynamic temporal encoding by on‐sensor reservoirs and simple deep neural network (DNN) blocks. The reservoir nodes enable efficient temporal processing of asynchronous events by leveraging the native dynamics of the node devices, while the DNN blocks enable spatial feature processing. Combining these blocks in a hierarchical structure, the RN‐Net offers efficient processing for both local and global spatiotemporal features. RN‐Net executes dynamic vision tasks created by event‐based cameras at the highest accuracy reported to date at one order of magnitude smaller network size. The use of simple DNN and standard backpropagation‐based training rules further reduces implementation and training costs.

Testing Dynamic Balance in People with Multiple Sclerosis: A Correlational Study between Standard Posturography and Robotic-Assistive Device.

Robotic devices are known to provide pivotal parameters to assess motor functions in Multiple Sclerosis (MS) as dynamic balance. However, there is still a lack of validation studies comparing innovative technologies with standard solutions. Thus, this study's aim was to compare the postural assessment of fifty people with MS (PwMS) during dynamic tasks performed with the gold standard EquiTest® and the robotic platform hunova®, using Center of Pressure (COP)-related parameters and global balance indexes. Pearson's ρ correlations were run for each COP-related measure and the global balance index was computed from EquiTest® and hunova® in both open (EO) and closed-eyes (EC) conditions. Considering COP-related parameters, all correlations were significant in both EO (0.337 ≤ ρ ≤ 0.653) and EC (0.344 ≤ ρ ≤ 0.668). Furthermore, Pearson's analysis of global balance indexes revealed relatively strong for visual and vestibular, and strong for somatosensory system associations (ρ = 0.573; ρ = 0.494; ρ = 0.710, respectively). Findings confirm the use of hunova® as a valid device for dynamic balance assessment in MS, suggesting that such a robotic platform could allow for a more sensitive assessment of balance over time, and thus a better evaluation of the effectiveness of personalized treatment, thereby improving evidence-based clinical practice.

Hybrid quantum physics-informed neural networks for simulating computational fluid dynamics in complex shapes

Finding the distribution of the velocities and pressures of a fluid by solving the Navier–Stokes equations is a principal task in the chemical, energy, and pharmaceutical industries, as well as in mechanical engineering and in design of pipeline systems. With existing solvers, such as OpenFOAM and Ansys, simulations of fluid dynamics in intricate geometries are computationally expensive and require re-simulation whenever the geometric parameters or the initial and boundary conditions are altered. Physics-informed neural networks (PINNs) are a promising tool for simulating fluid flows in complex geometries, as they can adapt to changes in the geometry and mesh definitions, allowing for generalization across fluid parameters and transfer learning across different shapes. We present a hybrid quantum PINN (HQPINN) that simulates laminar fluid flow in 3D Y-shaped mixers. Our approach combines the expressive power of a quantum model with the flexibility of a PINN, resulting in a 21% higher accuracy compared to a purely classical neural network. Our findings highlight the potential of machine learning approaches, and in particular HQPINN, for complex shape optimization tasks in computational fluid dynamics. By improving the accuracy of fluid simulations in complex geometries, our research using hybrid quantum models contributes to the development of more efficient and reliable fluid dynamics solvers.

Opto-magnonic reservoir computing coupling nonlinear interfered spin wave and visible light switching

Physical reservoir computing is a promising approach to realize high-performance artificial intelligence systems utilizing physical devices. Recently, it has been experimentally found that nonlinear interfered spin wave multi-detection shows excellent performance for processing nonlinear time-series data due to its outstanding features: nonlinearity, short-term memory, and the ability to map in high dimensional space. However, said performance is considerably inferior to reservoir computing utilizing an optical circuit with a large volume. Herein, we develop reservoir computing with nonlinear interfered spin wave coupled with light switching, namely opto-magnonic reservoir computing. The spin wave was modulated through a crystal field transition that occurred in two different Fe3+ sites of Y3Fe5O12 by visible light switching, and it was found that the spin wave modulated by visible light switching dramatically reduced normalized mean square errors to 4.96 × 10−3, 0.163, and 3.66 × 10−5 for NARMA2, NARMA10, and second-order nonlinear dynamical equation tasks. Said excellent performance results from the strong nonlinearity caused by chaos and large memory capacity induced by reservoir states diversified by visible light switching.

A deep reinforcement learning approach for dynamic task scheduling of flight tests

A deep reinforcement learning approach for dynamic task scheduling of flight tests

Augmenting complex and dynamic performance through mindfulness-based cognitive training: An evaluation of training adherence, trait mindfulness, personality and resting-state EEG.

Human performance applications of mindfulness-based training have demonstrated its utility in enhancing cognitive functioning. Previous studies have illustrated how these interventions can improve performance on traditional cognitive tests, however, little investigation has explored the extent to which mindfulness-based training can optimise performance in more dynamic and complex contexts. Further, from a neuroscientific perspective, the underlying mechanisms responsible for performance enhancements remain largely undescribed. With this in mind, the following study aimed to investigate how a short-term mindfulness intervention (one week) augments performance on a dynamic and complex task (target motion analyst task; TMA) in young, healthy adults (n = 40, age range = 18-38). Linear mixed effect modelling revealed that increased adherence to the web-based mindfulness-based training regime (ranging from 0-21 sessions) was associated with improved performance in the second testing session of the TMA task, controlling for baseline performance. Analyses of resting-state electroencephalographic (EEG) metrics demonstrated no change across testing sessions. Investigations of additional individual factors demonstrated that enhancements associated with training adherence remained relatively consistent across varying levels of participants' resting-state EEG metrics, personality measures (i.e., trait mindfulness, neuroticism, conscientiousness), self-reported enjoyment and timing of intervention adherence. Our results thus indicate that mindfulness-based cognitive training leads to performance enhancements in distantly related tasks, irrespective of several individual differences. We also revealed nuances in the magnitude of cognitive enhancements contingent on the timing of adherence, regardless of total volume of training. Overall, our findings suggest that mindfulness-based training could be used in a myriad of settings to elicit transferable performance enhancements.

Effect of a Short-Term Combined Balance and Multidirectional Plyometric Training on Postural Balance and Explosive Performance in U-13 Male and Female Soccer Athletes

This study’s aim is to examine the effect of a combined balance and multidirectional plyometric training intervention on postural balance ability and lower limb explosive performance in U-13 male and female soccer athletes. Twenty pre-adolescent (age: 12.6 ± 1.6 years) soccer athletes followed a 6-week training intervention combining balance exercises, dynamic stabilization tasks and multidirectional plyometric exercises at a frequency of twice/week for 20–25 min, based on a progressive increase in exercise difficulty from phase A (week 1–3) to phase B (week 4–6). Pre- and post-training measurements were carried out to assess the following: (a) static balance performance in single (left, right)-legged and two-legged quiet stance trials with eyes open and eyes closed (two trials per stance and vision condition of 30 s duration) and (b) lower limb explosive performance in countermovement and squat jumps without arm swing (three trials/jump). The vertical GRF was recorded by a customized force plate (Wii, 1.000 Hz, Biovision) and offline, CoP and explosive performance parameters were calculated. The overall results showed that the static balance ability of athletes remained unaffected, while restricting their vision deteriorated their postural control. The lower limb explosive performance showed a trend for improvement; however, inter-individual variations in athletes’ responses might have obscured any effect.

CrowdBot: An Open-Environment Robot Management System for On-Campus Services

In contemporary campus environments, the provision of timely and efficient services is increasingly challenging due to limitations in accessibility and the complexity and openness of the environment. Existing service robots, while operational, often struggle with adaptability and dynamic task management, leading to inefficiencies. To overcome these limitations, we introduce CrowdBot, a robot management system that enhances service in campus environments. Our system leverages a hierarchical reinforcement learning-based cloud-edge hybrid scheduling framework (REDIS), for efficient online streaming task assignment and dynamic action scheduling. To verify the REDIS framework, we have developed a digital twin simulation platform, which integrates large language models and hot-swapping technology. This facilitates seamless human-robot interaction, efficient task allocation, and cost-effective execution through the reuse of robot equipment. Our comprehensive simulations corroborate the system's remarkable efficacy, demonstrating significant improvements with a 24.46% reduction in task completion times, a 9.37% decrease in travel distances, and up to a 3% savings in power usage. Additionally, the system achieves a 7.95% increase in the number of tasks completed and a 9.49% reduction in response time. Real-world case studies further affirm CrowdBot's capability to adeptly execute tasks and judiciously recycle resources, thereby offering a smart and viable solution for the streamlined management of campus services.

An inversion-based clustering approach for complex clusters

BackgroundThe choice of an appropriate similarity measure plays a pivotal role in the effectiveness of clustering algorithms. However, many conventional measures rely solely on feature values to evaluate the similarity between objects to be clustered. Furthermore, the assumption of feature independence, while valid in certain scenarios, does not hold true for all real-world problems. Hence, considering alternative similarity measures that account for inter-dependencies among features can enhance the effectiveness of clustering in various applications.MethodsIn this paper, we present the Inv measure, a novel similarity measure founded on the concept of inversion. The Inv measure considers the significance of features, the values of all object features, and the feature values of other objects, leading to a comprehensive and precise evaluation of similarity. To assess the performance of our proposed clustering approach that incorporates the Inv measure, we evaluate it on simulated data using the adjusted Rand index.ResultsThe simulation results strongly indicate that inversion-based clustering outperforms other methods in scenarios where clusters are complex, i.e., apparently highly overlapped. This showcases the practicality and effectiveness of the proposed approach, making it a valuable choice for applications that involve complex clusters across various domains.ConclusionsThe inversion-based clustering approach may hold significant value in the healthcare industry, offering possible benefits in tasks like hospital ranking, treatment improvement, and high-risk patient identification. In social media analysis, it may prove valuable for trend detection, sentiment analysis, and user profiling. E-commerce may be able to utilize the approach for product recommendation and customer segmentation. The manufacturing sector may benefit from improved quality control, process optimization, and predictive maintenance. Additionally, the approach may be applied to traffic management and fleet optimization in the transportation domain. Its versatility and effectiveness make it a promising solution for diverse fields, providing valuable insights and optimization opportunities for complex and dynamic data analysis tasks.

Dynamic edge clustering and task scheduling for edge assisted metaverse system in the field of remote work and collaboration

SummaryThe metaverse, a future digital world for living, working, learning, and interacting, is rapidly gaining significance, particularly in the domain of remote work and collaboration. This emerging digital landscape demands high‐performance, low‐latency, and scalable services to provide an immersive user experience. The current technology used in metaverse systems has some limitations, which emphasize the importance of adopting emerging technologies like Edge Computing (EC). However, as the number of users and data volume increases, it can impact both system performance and scalability of the edge‐assisted metaverse system. Additionally, the uneven distribution of edge servers can cause inconsistencies and result in high latency. To overcome these challenges, this paper proposes a dynamic edge clustering and task scheduling approach for edge‐assisted metaverse systems (DTAM) in the field of remote work and collaboration. The proposed approach addresses the challenges of high user volume and uneven resource distribution by incorporating dynamic clustering and edge server assistance to improve clustering performance. Furthermore, a Prioritized Experience Replay‐based Deep Q‐learning algorithm with state augmentation (PERDQSA) for task scheduling is introduced to improve sample efficiency and performance. The performance of the proposed DTAM is evaluated against existing techniques, and experimental results demonstrate its significant superiority in terms of specific metrics such as bandwidth, task response time, energy efficiency, and latency. The experiments demonstrated that DTAM outperforms Transformation‐based Edge Computing Deep Q‐Learning (TransEC‐DQL) in several key metrics. Specifically, DTAM achieves 28.5% reduction in latency, 13.7% reduction in response time, and 6.4% improvement in bandwidth compared to TransEC‐DQL. These results signify that DTAM can deliver a significantly enhanced user experience in the metaverse, particularly in the context of remote work and collaboration.

An online integrated satellite–terrestrial IoT task offloading and service deployment strategy

As a core component of 6G networking, low earth orbit (LEO) satellite networks are considered a promising solution for providing network services in remote areas and have received considerable attention in recent years. Meanwhile, with the rapid development of IoT terminals and applications, an increasing emphasis has been placed on task offloading for real-time tasks. However, most existing works on task offloading for integrated satellite–terrestrial IoT (IST-IoT) applications are based on the assumption that the global task and network information is known, an assumption that is obviously not met in online real-time task offloading. In addition, current studies rarely consider the tradeoff among the task offloading volume, satellite endurance and wholesale cost of terrestrial resources, which are precisely the holistic issues that satellite network operators must confront. To address such issues, this study considers task offloading, service deployment and resource allocation in highly dynamic online task offloading scenarios with the goal of maximizing the average task offloading volume. Moreover, virtual queues are constructed in this study to account for the impacts of long-term satellite endurance and terrestrial network resource costs on task offloading. We first decouple the original problem to obtain a single-time-slot optimization problem within the Lyapunov optimization framework, and we then propose a 2-stage optimization approach based on deep reinforcement learning (DRL) to solve the complex single-time-slot mixed integer nonlinear programming (MINLP) problem. Simulation experiments demonstrate that our proposed algorithm can yield near-optimal solutions that show superior performance compared to existing results.

Mapping the defect acceptance for dynamically loaded components by exploiting DIC-based full-field receptances

Potential defect acceptance can be seen dependable upon the mapping of effective strains on the whole surface, due to dynamic loading of the components as they are mounted. With proper constitutive models and loading spectra, the experiment-based mapping of the equivalent stresses can be achieved from optical full-field receptances in an extremely dense grid, also for lightweight parts, without inertia distortions thanks to contactless measurements. Fatigue spectral methods turn this mapping into components’ life distributions, for a clear assessment of the material’s utilisation: a risk grading mapping for potential defects can be formulated in the area of inquiry in order to discriminate between safe and dangerous locations. By following this experiment-based approach, potential defects in exercise and production might be tolerated in safer locations, under the chosen dynamic task, with great savings in costs and maintenance. Among the image-based full-field measurement techniques, Hi-Speed DIC has proved to work in many environments, to be able to estimate full-field receptances of real components in their effective assembling and loading conditions, in different cases of industrial interest. The quality achieved by DIC in the receptance maps helps in numerically deriving the strain FRFs on the sensed surface. Extended scenarios – with newly modelled coloured noises for an advanced excitation definition from two shakers – and a vibrating rectangular plate, as real mounted component, are highlighted in details to prove the effectiveness of DIC-based risk index mapping under defect acceptance criteria.

Combining Reinforcement Learning and Tensor Networks, with an Application to Dynamical Large Deviations.

We present a framework to integrate tensor network (TN) methods with reinforcement learning (RL) for solving dynamical optimization tasks. We consider the RL actor-critic method, a model-free approach for solving RL problems, and introduce TNs as the approximators for its policy and value functions. Our "actor-critic with tensor networks" (ACTeN) method is especially well suited to problems with large and factorizable state and action spaces. As an illustration of the applicability of ACTeN we solve the exponentially hard task of sampling rare trajectories in two paradigmatic stochastic models, the East model of glasses and the asymmetric simple exclusion process, the latter being particularly challenging to other methods due to the absence of detailed balance. With substantial potential for further integration with the vast array of existing RL methods, the approach introduced here is promising both for applications in physics and to multi-agent RL problems more generally.

Ontology based autonomous robot task processing framework.

In recent years, the perceptual capabilities of robots have been significantly enhanced. However, the task execution of the robots still lacks adaptive capabilities in unstructured and dynamic environments. In this paper, we propose an ontology based autonomous robot task processing framework (ARTProF), to improve the robot's adaptability within unstructured and dynamic environments. ARTProF unifies ontological knowledge representation, reasoning, and autonomous task planning and execution into a single framework. The interface between the knowledge base and neural network-based object detection is first introduced in ARTProF to improve the robot's perception capabilities. A knowledge-driven manipulation operator based on Robot Operating System (ROS) is then designed to facilitate the interaction between the knowledge base and the robot's primitive actions. Additionally, an operation similarity model is proposed to endow the robot with the ability to generalize to novel objects. Finally, a dynamic task planning algorithm, leveraging ontological knowledge, equips the robot with adaptability to execute tasks in unstructured and dynamic environments. Experimental results on real-world scenarios and simulations demonstrate the effectiveness and efficiency of the proposed ARTProF framework. In future work, we will focus on refining the ARTProF framework by integrating neurosymbolic inference.

A Dynamic Tasking-Based Evolutionary Algorithm for Bi-Objective Feature Selection

Feature selection in classification is a complex optimization problem that cannot be solved in polynomial time. Bi-objective feature selection, aiming to minimize both selected features and classification errors, is challenging due to the conflict between objectives, while one of the most effective ways to tackle this is to use multi-objective evolutionary algorithms. However, very few of these have ever reflected an evolutionary multi-tasking framework, despite the implicit parallelism offered by the population-based search characteristic. In this paper, a dynamic multi-tasking-based multi-objective evolutionary algorithm (termed DTEA) is proposed for handling bi-objective feature selection in classification, which is not only suitable for datasets with relatively lower dimensionality of features, but is also suitable for datasets with relatively higher dimensionality of features. The role and influence of multi-tasking on multi-objective evolutionary feature selection were studied, and a dynamic tasking mechanism is proposed to self-adaptively assign multiple evolutionary search tasks by intermittently analyzing the population behaviors. The efficacy of DTEA is tested on 20 classification datasets and compared with seven state-of-the-art evolutionary algorithms. A component contribution analysis was also conducted by comparing DTEA with its three variants. The empirical results show that the dynamic-tasking mechanism works efficiently and enables DTEA to outperform other algorithms on most datasets in terms of both optimization and classification.

Bilingualism Predicts Affective Theory of Mind in Autistic Adults.

This study examined the impact of bilingualism on affective theory of mind (ToM) and social prioritization (SP) among autistic adults compared to neurotypical comparison participants. Fifty-two (25 autistic, 27 neurotypical) adult participants (ages 21-35 years) with varying second language (L2) experience, ranging from monolingual to bilingual, completed an affective ToM task. A subset of this sample also completed a dynamic eye-tracking task designed to capture differences in time spent looking at social aspects of a scene (SP). Four language groups were compared on task performance (monolingual autism and neurotypical, bilingual autism and neurotypical), followed by analyses examining the contribution of L2 experience, autism characteristics, and social face prioritization on affective ToM, controlling for verbal IQ. Finally, we conducted an analysis to identify the contribution of SP on affective ToM when moderated by autism status and L2 experience, controlling for verbal IQ. The monolingual autism group performed significantly worse than the other three groups (bilingual autism, monolingual neurotypical, and bilingual neurotypical) on the affective ToM task; however, there were no significant differences between the bilingual autism group compared to the monolingual and bilingual neurotypical groups. For autistic individuals, affective ToM capabilities were positively associated with both verbal IQ and L2 experience but did not relate to autism characteristics or SP during eye tracking. Neurotypical participants showed greater SP during the eye-tracking task, and SP did not relate to L2 or autism characteristics for autistic individuals. SP and verbal IQ predicted affective ToM performance across autism and neurotypical groups, but this relationship was moderated by L2 experience; SP more strongly predicted affective ToM performance among participants with lower L2 experience (e.g., monolingual) and had less of an impact for those with higher L2 experience. This study provides support for a bilingual advantage in affective ToM for autistic individuals. https://doi.org/10.23641/asha.25696083.

Deep Reinforcement Learning techniques for dynamic task offloading in the 5G edge-cloud continuum

The integration of new Internet of Things (IoT) applications and services heavily relies on task offloading to external devices due to the constrained computing and battery resources of IoT devices. Up to now, Cloud Computing (CC) paradigm has been a good approach for tasks where latency is not critical, but it is not useful when latency matters, so Multi-access Edge Computing (MEC) can be of use. In this work, we propose a distributed Deep Reinforcement Learning (DRL) tool to optimize the binary task offloading decision, this is, the independent decision of where to execute each computing task, depending on many factors. The optimization goal in this work is to maximize the Quality-of-Experience (QoE) when performing tasks, which is defined as a metric related to the battery level of the UE, but subject to satisfying tasks’ latency requirements. This distributed DRL approach, specifically an Actor-Critic (AC) algorithm running on each User Equipment (UE), is evaluated through the simulation of two distinct scenarios and outperforms other analyzed baselines in terms of QoE values and/or energy consumption in dynamic environments, also demonstrating that decisions need to be adapted to the environment’s evolution.

Dynamic heterogeneous federated learning with multi-level prototypes

Federated learning shows promise as a privacy-preserving collaborative learning technique. Existing research mainly focuses on skewing the class distribution across clients. However, most approaches suffer from catastrophic forgetting and classifier shift, mainly when the global distribution of all classes is extremely unbalanced and the data distribution of the client dynamically evolves over time. In this paper, we study the Dynamic Heterogeneous Federated Learning, which addresses the practical scenario where heterogeneous data distributions exist among different clients and dynamic tasks within the client. Accordingly, we propose a novel federated learning framework named Federated Multi-Level Prototypes and design federated multi-level regularizations. To mitigate classifier shift, we construct semantic prototypes to provide fruitful generalization knowledge. To maintain the model stability and consistency convergence, three regularizations are introduced as training losses, i.e., prototype-based regularization, semantic prototype-based regularization, and federated inter-task regularization. Extensive experiments show that the proposed method achieves advanced performance in various settings.