Satellite networks have become vital for Earth observation, environmental monitoring, and disaster warning by interconnecting remote sensing satellites, drones, and ground stations into large-scale distributed monitoring systems. Federated learning (FL) enables intelligent applications on these systems while preserving data privacy. However, satellite networks face challenges such as high latency and jitter, which hinder FL training efficiency and model timeliness. To address these issues, we propose a three-layer asynchronous FL framework that integrates terminals, Low Earth Orbit (LEO), and Medium Earth Orbit (MEO). This architecture employs asynchronous updates, distributed partial aggregation, and hierarchical global aggregation to enhance model convergence under unstable communication conditions. We introduce a dynamic weighted asynchronous update mechanism that mitigates the impact of stale updates based on staleness and similarity metrics. Additionally, a reinforcement learning-based adaptive timeslot request strategy is designed, allowing terminals to optimize global model update timings via a teacher-student model, thus reducing staleness caused by delay jitter. Simulation results on a multilayer satellite network demonstrate that our method significantly improves convergence speed and model accuracy. Compared to baseline methods, dynamic weighting enhances accuracy by 1.12%–1.80%, while adaptive timeslot requests improve accuracy by 3.83%–6.42% and reduce convergence time by 25.79%–59.89%.

The interaction between negative emotion contagion and the spread of extreme opinions on social media can lead to serious consequences, highlighting the need to understand the co-evolution of opinion and emotion. However, most existing studies treat opinion and emotion dynamics as independent processes. The few models that explore their interplay often do not fully capture their distinct dynamic mechanisms and tend to overlook the mutual reinforcement between opinion convergence and emotion contagion. To address these limitations, we propose the dynamics of opinion and emotion on social media (DOES) model, which employs two differential equations to represent the separate dynamics of opinion and emotion, and introduces coupling functions that describe their mutual reinforcement. We analyze the model’s steady-state conditions and validate the results through simulations. Prediction experiments on real-world datasets demonstrate the model’s ability to reproduce key patterns. DOES provides a novel framework for modeling the coevolution of opinion and emotion, offering insights into emotional regulation and discourse moderation on social media platforms.

Digital twin is emerging as a key technology to monitor the status of complex industry systems. Valuable insights, such as running statuses and anomalies, can be analyzed from the collected system status timely. Considering that the data updating from each system component (known as a physical object) to its digital twin is performed continuously, timely and accurate streaming analytics based on machine learning models is a key technology to analyze such data efficiently. In this paper, we focus on enabling low-delay yet highly-accurate streaming analytics for digital twin applications in mobile edge computing (MEC) networks. Specifically, we formulate a fundamental optimization problem of digital twin placements and model selections for streaming analytics, with the aim of minimizing both the analytic loss and the processing delay. To this end, we first consider the problem with a single query, for which, we propose an approximation algorithm with provable approximation ratio for a special case, and then devise an efficient algorithm for the original problem with a single query. We then study the online digital twin placement and model selection problem for streaming analytics with multiple queries under real scenarios, where resource demands of arrival queries and resource availability of MEC network are uncertain. We propose an online learning algorithm with a bounded regret to make admission policies. We finally evaluate the performance of the proposed algorithms by extensive simulations. Results show that the weighted sums of the total processing delay and the cumulative loss in the solution delivered by the proposed algorithms outperform their counterparts by 12.5% with a single query and 13.3% with multiple queries, respectively.

Data stream processing is a functionality of various AI applications to obtain continuous insights from data streams. Serverless edge computing (SEC) is a key solution for implementing data stream processing requests by deploying serverless functions into cloudlets. However, existing data stream processing methods focus more on processing delay, ignoring fault tolerance and complex dependencies among functions, resulting in critical events being missed in the event of any fault and processing inefficiency. Besides, due to the uncertainty of data streams, existing function deployment methods may not be suitable for their newly changed data rates, causing resource waste or shortages. To address these problems, we first propose an optimization framework to enable efficient and fault tolerant function deployment, such that the delay of data stream processing is minimized while meeting its fault tolerant requirements and resource capacity constraints of cloudlets in an SEC network. We then design an online learning algorithm that predicts data rate changes through a multi-timescale machine learning method and proactively adjusts instance locations and numbers to absorb data rate uncertainty. Experimental results in a real test-bed show that our proposed algorithms outperform their counterparts by 13.5% on the average delay and 26.3% on the average fault tolerance.

Serverless computing are invented to free developers of analytic services from the management of cloud resources. Developers only need to submit their code to a serverless edge cloud (SEC) and the cloud platform matches the submitted code to itsserverless functions, enabled by the paradigm of Function as a Service (FaaS). However, serverless functions usually are short-lived with limited resources. In this paper, we aim to address the challenge of how to fully utilize the short-lived serverless functions to enable continuous analysis of the most freshly-generated big data. We first formulate an optimization problem of age-aware big data query evaluation in an SEC network so that theage of datais minimized, where the age of data is the duration between the generation time of the data and the current time. We then propose approximation algorithms for the age-aware big data query evaluation problem of a single query, by proposing a parameterized virtualization technique that smartly handles the large resource demands of big data queries in short-lived and resource-constraint serverless functions. We further consider the scenario where big data queries arrive into the system one-by-one and their resource demands are uncertain. To this end, we devise an online learning algorithm with bounded regret by leveraging a memory repacking mechanism to improve resource utilization, for the problem of online age-aware big data query evaluation. We evaluate the performance of the proposed algorithms through extensive simulations to testify their performances in large scales. We also validate the effectiveness of the proposed mechanism in a real test-bed built on Kubernetes and OpenWhisk with TPC-DS benchmarks. Experimental results show that the proposed algorithms outperform the state-of-the-arts, by reducing the age of data by$8.82\%$on average.

Predicting roll call votes through modeling political actors has emerged as a focus in quantitative political science and computer science. Widely used embedding-based methods generate vectors for legislators from diverse data sets to predict legislative behaviors. However, these methods often contend with challenges such as the need for manually predefined features, reliance on extensive training data, and a lack of interpretability. Achieving more interpretable predictions under flexible conditions remains an unresolved issue. This paper introduces the Political Actor Agent (PAA), a novel agent-based framework that utilizes Large Language Models to overcome these limitations. By employing role-playing architectures and simulating legislative system, PAA provides a scalable and interpretable paradigm for predicting roll-call votes. Our approach not only enhances the accuracy of predictions but also offers multi-view, human-understandable decision reasoning, providing new insights into political actor behaviors. We conducted comprehensive experiments using voting records from the 117-118th U.S. House of Representatives, validating the superior performance and interpretability of PAA. This study not only demonstrates PAA's effectiveness but also its potential in political science research.

Dynamic graphs evolve continuously, presenting challenges for traditional graph learning due to their changing structures and temporal dependencies. Recent advancements have shown potential in addressing these challenges by developing suitable meta-learning-based dynamic graph neural network models. However, most meta-learning approaches for dynamic graphs rely on fixed weight update parameters, neglecting the essential intrinsic complex high-order topological information of dynamically evolving graphs. We have designed Dowker Zigzag Persistence (DZP), an efficient and stable dynamic graph persistent homology representation method based on Dowker complex and zigzag persistence, to capture the high-order features of dynamic graphs. Armed with the DZP ideas, we propose TMetaNet, a new meta-learning parameter update model based on dynamic topological features. By utilizing the distances between high-order topological features, TMetaNet enables more effective adaptation across snapshots. Experiments on real-world datasets demonstrate TMetaNet’s state-of-the-art performance and resilience to graph noise, illustrating its high potential for meta-learning and dynamic graph analysis. Our code is available at https://github.com/Lihaogx/TMetaNet.

Polarization and fragmentation in social media amplify user biases, making it increasingly important to understand the evolution of opinions. Opinion dynamics provide interpretability for studying opinion evolution, yet incorporating these insights into predictive models remains challenging. This challenge arises due to the inherent complexity of the diversity of opinion fusion rules and the difficulty in capturing equilibrium states while avoiding over-smoothing. This paper constructs a unified opinion dynamics model to integrate different opinion fusion rules and generates corresponding synthetic datasets. To fully leverage the advantages of unified opinion dynamics, we introduces UniGO, a framework for modeling opinion evolution on graphs. Using a coarsen-refine mechanism, UniGO efficiently models opinion dynamics through a graph neural network, mitigating over-smoothing while preserving equilibrium phenomena. UniGO leverages pretraining on synthetic datasets, which enhances its ability to generalize to real-world scenarios, providing a viable paradigm for applications of opinion dynamics. Experimental results on both synthetic and real-world datasets demonstrate UniGO's effectiveness in capturing complex opinion formation processes and predicting future evolution. The pretrained model also shows strong generalization capability, validating the benefits of using synthetic data to boost real-world performance.

Persistent homology, a fundamental technique within Topological Data Analysis (TDA), captures structural and shape characteristics of graphs, yet encounters computational difficulties when applied to dynamic directed graphs. This paper introduces the Dynamic Neural Dowker Network (DNDN), a novel framework specifically designed to approximate the results of dynamic Dowker filtration, aiming to capture the high-order topological features of dynamic directed graphs. Our approach creatively uses line graph transformations to produce both source and sink line graphs, highlighting the shared neighbor structures that Dowker complexes focus on. The DNDN incorporates a Source-Sink Line Graph Neural Network (SSLGNN) layer to effectively capture the neighborhood relationships among dynamic edges. Additionally, we introduce an innovative duality edge fusion mechanism, ensuring that the results for both the sink and source line graphs adhere to the duality principle intrinsic to Dowker complexes. Our approach is validated through comprehensive experiments on real-world datasets, demonstrating DNDN's capability not only to effectively approximate dynamic Dowker filtration results but also to perform exceptionally in dynamic graph classification tasks.

Low Earth Orbit (LEO) satellite networks provide global connectivity but are vulnerable to security threats such as link flooding attacks. To defend against such attacks, stateof-the-art approaches employ SDN to acquire a global view of the network, enabling the detection and mitigation of malicious traffic. However, in LEO constellation networks, the distributed nature of satellites across a large spatial scale introduces significant latency in both satellite-to-ground and inter-satellite links, with latency reaching up to tens of milliseconds, while attack traffic dynamically adapts within sub-milliseconds. As a result, existing defense systems face challenges in countering these attacks effectively due to the increased reaction time caused by link latency. In this paper, we leverage programmable switches to build a real-time defense system against link flooding attacks (LFA) in LEO constellation networks. To achieve this, we analyze the practical constraints encountered in the deployment of LFA attacks against state-of-the-art LEO satellite systems. We observe that despite the ability of bots to initiate attack traffic from any location worldwide, an anomalous distribution of flow rate on the affected links can still be detected. We propose SatShield, an in-network defense system that filters out suspicious traffic (heavy flows) in the network and mitigates these threats by leveraging programmable packet scheduling. By using SatShield, we are able to achieve real-time identification and rate-limiting of attacks at line rate on a per-packet basis. We implement SatShield with P4 in a commercial programmable switch and evaluate it with real-world traffic traces. Our evaluation shows that SatShield autonomously identifies LFA attack flows and rapidly mitigates LFA attacks.

To unveil the hidden value in the datasets of user equipments (UEs) while preserving user privacy, federated learning (FL) is emerging as a promising technique to train a machine learning model using the datasets of UEs locally without uploading the datasets to a central location. Customers require to train machine learning models based on different datasets of UEs, through issuing FL requests that are implemented by FL services in a mobile edge computing (MEC) network. A key challenge of enabling FL in MEC networks is how to minimize the energy consumption of implementing FL requests while guaranteeing the accuracy of machine learning models, given that the availabilities of UEs usually are uncertain. In this paper, we investigate the problem of energy minimization for FL in an MEC network with uncertain availabilities of UEs. We first consider the energy minimization problem for a single FL request in an MEC network. We then propose a novel optimization framework for the problem with a single FL request, which consists of (1) an online learning algorithm with a bounded regret for the UE selection, by considering various contexts (side information) that influence energy consumption; and (2) an approximation algorithm with an approximation ratio for the aggregator placement for a single FL request. We third deal with the problem with multiple FL requests, for which we devise an online learning algorithm with a bounded regret. We finally evaluate the performance of the proposed algorithms by extensive experiments. Experimental results show that the proposed algorithms outperform their counterparts by reducing at least 13% of the total energy consumption while achieving the same accuracy.

This paper focuses on the mobility prediction problem of specific occupational groups that rely on mobile devices, predicting the mobilities of these groups using data shared across different platforms. While the mobility patterns of general populations have been studied extensively, predicting the movements of specific occupational groups like ride-hailing drivers, who frequently move over long distances, requires a more nuanced approach. This paper introduces FedOGM, a federated learning framework designed to predict the mobility of specific occupational groups while preserving user privacy. The framework utilizes a dynamic bidirectional graph attention network model DyBGAT for predicting the edge weights of an occupational dynamic Origin-Destination graph, representing the mobility behavior. In order to address key challenges in federated learning, FedOGM integrates By leveraging multi-source data from users’ mobile devices and extracting unique occupational features and occupational features into the framework, such as applications usage duration feature, location switching feature, and call duration feature. Assisted by these personalized occupational features, we have devised a client selection algorithm, generated occupational dynamic OD graphs to diminish communication overhead, and proposed a group-level FedAvg aggregation method. Experiments conducted on real-world datasets of ride-hailing drivers, ride-hailing truck drivers, and real estate salespersons, including scenarios with false data, confirm the efficacy of the proposed methods in both mobility prediction and privacy protection.

Federated learning (FL) is a distributed machine learning technique that enables model development on user equipments (UEs) locally, without violating their data privacy requirements. Conventional FL adopts a single parameter server to aggregate local models from UEs, and can suffer from efficiency and reliability issues – especially when multiple users issue concurrentFL requests. Hierarchical FL consisting of a master aggregator and multiple worker aggregators to collectively combine trained local models from UEs is emerging as a solution to efficient and reliable FL. The placement of worker aggregators and assignment of UEs to worker aggregators plays a vital role in minimizing the cost of implementing FL requests in a Mobile Edge Computing (MEC) network. Cost minimization associated with joint worker aggregator placement and UE assignment problem in an MEC network is investigated in this work. An optimization framework for FL and an approximation algorithm with an approximation ratio for a single FL request is proposed. Online worker aggregator placements and UE assignments for dynamic FL request admissions with uncertain neural network models, where FL requests arrive one by one without the knowledge of future arrivals, is also investigated by proposing an online learning algorithm with a bounded regret. The performance of the proposed algorithms is evaluated using both simulations and experiments in a real testbed with its hardware consisting of server edge servers and devices and software built upon an open source hierarchical FedML (HierFedML) environment. Simulation results show that the performance of the proposed algorithms outperform their benchmark counterparts, by reducing the implementation cost by at least 15% per FL request. Experimental results in the testbed demonstrate the performance gain using the proposed algorithms using real datasets for image identification and text recognition applications.