Multi-agent reinforcement learning provides promising prospect for task scheduling in cloud-edge computing environment in recent years. However, there remains a formidable challenge due to partial observation and the rigid coupling between action spaces and schedulable devices. These limit the ability of agent to perceive global communication patterns and adapt to dynamic environments, resulting in unsatisfactory scheduling decisions. To address these issues, this work proposes SPAD, a novel spatial perception and action decoupling empowered distributed multi-agent AI task scheduling framework. By constructing a global spatial feature distillation mechanism, SPAD can approximate the implicit heterogeneous connection patterns and communication dynamics between devices and tasks under constrained observability, enhancing its ability to make robust decisions in dynamic environments with limited observations. Additionally, SPAD employs a Lyapunov-based action decoupling module to alleviate scalability challenges from rigid action-device coupling, while a novel intrinsic penalty mechanism augments the agent’s advantage function with the instantaneous Lyapunov cost, thereby aligning the policy optimization process with the decoupling module’s underlying stability constraints. Through a comprehensive empirical evaluation spanning synthetic, bursty, and real-world trace-driven workloads, we show that SPAD consistently outperforms state-of-the-art benchmarks in reducing task completion latency and improving resource utilization, while maintaining remarkable resilience and scalability across diverse network topologies and under non-stationary load conditions.

In recent years, China's economy has developed rapidly, but the relationship between environmental protection and sustainable economic development has attracted increasing attention. Based on macro data, this study uses principal component analysis method, selects Beijing economic and environmental quality data, uses regression model to fit EKC curve, and describes the causal relationship between environmental quality and economic sustainable development through regression analysis. The experimental results show that the relationship between economy and environment in Beijing presents an “inverted N-shaped curve”, which is not suitable for the traditional Kuznets “inverted U-shaped” curve. This study is of great practical significance to the coordinated development of regional environment and economy. This study can promote the understanding and exchange of related technologies and knowledge theories between academic circles and environmental, economic and other organizations, thus completing the transfer and sharing of knowledge.

This article proposes a Kolmogorov Arnold Networks (KANs) model that combines B-spline functions and is applied to a multimodal data fusion defense strategy for campus network security. Simulation experiments analyze the optimization effect of KANs models in various scenarios, including factors like network layers, adaptive learning rate, early stopping, L2 regularization, weighted loss function, and batch normalization. Experimental results show that, after optimizing the network structure and adopting adaptive learning rates, the KANs model converges more quickly and avoids overfitting. Especially with the introduction of early stopping strategy, L2 regularization, and Batch Normalization, the generalization ability of the model has been significantly improved. The strategy based on multimodal data fusion effectively improves the robustness and accuracy of the model when processing different types of data, such as images, text, time series data, etc. Experiments have shown that through weighted loss functions and multi task learning strategies, the model exhibits higher accuracy and lower false alarm rates when dealing with imbalanced samples and complex attack patterns. Overall, the proposed KANs model, combined with B-spline functions, not only optimizes campus network security defense but also improves the accuracy and robustness of multimodal data fusion tasks.

Malaria is one of the deadliest infectious diseases globally, claiming hundreds of thousands of lives each year. The disease presents substantial heterogeneity among the population, with approximately two-thirds of fatalities occurring in children under five years old. Immunity to malaria develops through repeated exposure and plays a crucial role in disease dynamics. Seasonal environmental fluctuations, such as changes in temperature and rainfall, lead to temporal heterogeneity and further complicate transmission dynamics and the utility of intervention strategies. We employ an age-structured partial differential equation model to characterize seasonal malaria transmission and assess vaccination strategies that vary by timing and duration. Our model integrates vector-host epidemiological dynamics across different age groups and nonlinear feedback between transmission and immunity. We calibrate the model to year-round and seasonal malaria settings and conduct extensive sensitivity analyses for both scenarios to systematically assess which assumptions lead to the most uncertainty. We use time-varying sensitivity indices to identify critical disease parameters during low and high transmission seasons. We further investigate the impact of vaccination and its implementation in the seasonal malaria settings. When implementing a three-dose primary vaccination series, seasonally targeted campaigns can prevent significantly more cases per vaccination than constant year-long programs in regions with strong seasonal variation in transmission. In such scenarios, the optimal vaccination interval aligns with the peak in infected mosquito abundance and precedes the peak in malaria transmission. In contrast, seasonal booster programs may provide limited advantages over year-long vaccination. Additionally, while increasing annual vaccination counts can reduce overall disease incidence, it yields marginal improvements in cases prevented per vaccination.

Studies of resource provision in cloud computing have drawn extensive attention, since effective task scheduling solutions promise an energy-efficient way of utilizing resources while meeting diverse requirements of users. Deep reinforcement learning (DRL) has demonstrated its outstanding capability in tackling this issue with the ability of online self-learning, however, it is still prevented by the low sampling efficiency, poor sample validity, and slow convergence speed especially for deadline constrained applications. To address these challenges, an Imitation Learning Enabled Fast and Adaptive Task Scheduling (ILETS) framework based on DRL is proposed in this paper. First, we introduce behavior cloning to provide a well-behaved and robust model through Offline Initial Network Parameters Training (OINPT) so as to guarantee the initial decision-making quality of DRL. Next, we design a novel Online Asynchronous Imitation Learning (OAIL)-based method to assist the DRL agent to re-optimize its policy and to against the oscillations caused by the high dynamic of the cloud, which promises DRL agent moving towards the optimal policy with a fast and stable process. Extensive experiments on the real-world dataset have demonstrated that the proposed ILETS can consistently produce shorter response time , lower energy consumption and higher success rate than the baselines and other state-of-the-art methods at the accelerated convergence speed.

Managing scientific applications in the Cloud poses many challenges in terms of workflow scheduling, especially in handling multi-objective workflow scheduling under quality of service (QoS) constraints. However, most studies address the workflow scheduling problem on the premise of the unchanged environment, without considering the high dynamics of the Cloud. In this paper, we model the constrained workflow scheduling in a dynamic Cloud environment as a dynamic multi-objective optimization problem with preferences, and propose a transfer learning based multi-objective evolutionary algorithm (TL-MOEA) to tackle the workflow scheduling problem of dynamic nature. Specifically, an elite-led transfer learning strategy is proposed to explore effective parameter adaptation for the MOEA by transferring helpful knowledge from elite solutions in the past environment to accelerate the optimization process. In addition, a multi-space diversity learning strategy is developed to maintain the diversity of the population. To satisfy various QoS constraints of workflow scheduling, a preference-based selection strategy is further designed to enable promising solutions for each iteration. Extensive experiments on five well-known scientific workflows demonstrate that TL-MOEA can achieve highly competitive performance compared to several state-of-art algorithms, and can obtain triple win solutions with optimization objectives of minimizing makespan, cost and energy consumption for dynamic workflow scheduling with user-defined constraints.

Edge sensing can achieve high-performance state estimation in industrial IoT systems by supporting task offloading and data processing at powerful edge estimators. Accurate edge sensing depends on low offloading delay. However, it is challenging to decrease offloading delay due to the harsh industrial environment and limited communication-and-computation resources. In this article, a closed-form expressing of estimation error with respect to offloading delay is derived to indicate that adjusting offload delay on demand is necessary for estimation error reduction. Then, we propose an adaptive edge sensing scheme, aiming to minimize estimation error by jointly optimizing task offloading and sensor scheduling. The required optimization is formulated as a mixed-integer nonlinear programming problem and solved by the designed decomposition and approximation methods. Specifically, the maximum matching is used for sensor scheduling to assign the optimal edge estimator for each sensor. The task offloading algorithm is designed based on the inner approximation method to reduce the offloading delay. Finally, simulation results demonstrate that the proposed scheme has superiorities in reducing estimation error compared with centralized sensing and distributed sensing schemes. Moreover, we find an interesting result that estimation error is delay sensitive when the offloading delay is large.

The development of remaining oil plays an important role in increasing the late production, and low-grade faults seriously impacts residual oil exploration and development. Low-grade faults have a small fault displacement, brief extension, and strong concealment, which hinders their prediction using the typical semantic segmentation network. To intelligently identify low-grade faults, we designed a codec target edge detection technique. For the network to fully learn the low-grade faults information, we constructed the encoder using dilated convolution. Next, we introduced an attention mechanism to the decoder to improve capture of location information from a shallow network and semantic information from a deep network. Finally, the multi-scale fusion decoder outputs fault information of different scales, which further improves the identification accuracy of low-grade faults. The training model is applied to simulated data and actual seismic data through ablation experiments. The results show that this method can effectively identify low-grade faults and overcomes problems such as blurred fault cross-location, thicker edge contour lines, lower detection accuracy, and less training data. Compared with conventional a Holistically-Nested Edge Detection (HED) and a semantic segmentation (UNet), fault misidentification is reduced, fault continuity is increased, and fault accuracy is improved. providing technical support for the exploration and development of remaining oil and increasing the recovery rate of old oil fields.

MOTIVATION: Single-cell RNA sequencing (scRNAseq) technologies allow for measurements of gene expression at a single-cell resolution. This provides researchers with a tremendous advantage for detecting heterogeneity, delineating cellular maps or identifying rare subpopulations. However, a critical complication remains: the low number of single-cell observations due to limitations by rarity of subpopulation, tissue degradation or cost. This absence of sufficient data may cause inaccuracy or irreproducibility of downstream analysis. In this work, we present Automated Cell-Type-informed Introspective Variational Autoencoder (ACTIVA): a novel framework for generating realistic synthetic data using a single-stream adversarial variational autoencoder conditioned with cell-type information. Within a single framework, ACTIVA can enlarge existing datasets and generate specific subpopulations on demand, as opposed to two separate models [such as single-cell GAN (scGAN) and conditional scGAN (cscGAN)]. Data generation and augmentation with ACTIVA can enhance scRNAseq pipelines and analysis, such as benchmarking new algorithms, studying the accuracy of classifiers and detecting marker genes. ACTIVA will facilitate analysis of smaller datasets, potentially reducing the number of patients and animals necessary in initial studies. RESULTS: We train and evaluate models on multiple public scRNAseq datasets. In comparison to GAN-based models (scGAN and cscGAN), we demonstrate that ACTIVA generates cells that are more realistic and harder for classifiers to identify as synthetic which also have better pair-wise correlation between genes. Data augmentation with ACTIVA significantly improves classification of rare subtypes (more than 45% improvement compared with not augmenting and 4% better than cscGAN) all while reducing run-time by an order of magnitude in comparison to both models. AVAILABILITY AND IMPLEMENTATION: The codes and datasets are hosted on Zenodo (https://doi.org/10.5281/zenodo.5879639). Tutorials are available at https://github.com/SindiLab/ACTIVA. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

News comment is a new text genre which people use to express their opinions on recent news events. Different from normal text corpus, news comments have some particular properties. The named entities in the news comments usually use some wrongly written words, informal abbreviations or aliases, which bring great difficulties for machine detection and understanding. This paper addresses the issue of named entity resolution in Chinese news comments on the web, which is a special case of coreference resolution. Traditional resolution algorithms have some limitations for this special task. In this paper, we first define the special task, and then propose a novel resolution algorithm with new features to improve the resolution performance. We manually labeled a benchmark dataset with 60 pieces of news and their corresponding comments downloaded from a popular Chinese news portal and the experimental results on the dataset show that our algorithm is effective for this special task.