The ability of T-cell receptors (TCRs) to recognize neoantigens is fundamental to the initiation and maintenance of adaptive immune responses. In TCR-based immunotherapies, elucidating the recognition patterns of TCRs for peptides and accurately identifying therapeutically relevant TCR-peptide pairs remain critical challenges. Here, we present a novel dual-pathway network model, ProTCR, which integrates the protein language model ProtT5 with deep learning methods. By incorporating both global and local feature extraction mechanisms, ProTCR enables efficient representation of amino acid sequences, thereby enhancing the model's generalizability across diverse data distributions and improving its biological interpretability. ProTCR demonstrates robust performance and broad applicability across various datasets, including neoantigens, previously unseen peptides, and MHC class II-restricted epitopes, overcoming the reliance on known peptide-TCR pairs observed in previous studies. It also offers new insights for predicting diverse classes of antigenic peptides. We applied ProTCR to several clinically relevant scenarios, including immunotherapeutic target identification in acute myeloid leukemia, neoantigen-targeted immunotherapy in solid tumours, and antigen-specific T cell recognition against pathogens such as influenza and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Across these complex settings, ProTCR consistently maintained high accuracy and stability, demonstrating strong cross-task adaptability and broad potential for clinical application. This work not only provides a powerful tool for elucidating immune response mechanisms but also offers a solid computational foundation for the design of neoantigen or TCR based precision immunotherapy strategies.

In future 6G networks, mobile edge computing (MEC) is envisioned to offer integrated computing, communication, and storage services at the network edge, enhancing both computational efficiency and communication quality. However, most existing MEC designs neglect the impact of jamming attacks, especially those from intelligent and adaptive jammers. To fill this important research gap, this paper investigates a jamming-resilient MEC framework that aims to improve communication reliability and reduce system delay under adversarial interference. Leveraging the emerging movable antenna (MA) technology, which allows dynamic adjustment of antenna positions and orientations, we propose a novel MA-assisted anti-jamming MEC architecture. Unlike existing works, our model explicitly considers the movement delay caused by MA, which is critical for practical deployment. We jointly optimize the MA positions at both the user equipment (UE) and the base station (BS), BS transmit beamforming, and task offloading ratios to minimize the total system delay. The resulting optimization problem is non-convex and highly coupled. Thus, we develop an efficient algorithm based on penalty dual decomposition (PDD) and successive convex approximation (SCA). Simulation results demonstrate that the proposed scheme significantly outperforms traditional fixed-position antenna (FPA) baselines in terms of jamming resilience and delay minimization, offering new insights into robust MEC system design for 6G networks.

Vehicular metaverses enable immersive digital experiences through seamless Vehicle Twin (VT) services. As vehicles move, VT service instances must migrate between RoadSide Units (RSUs) to sustain low-latency interactions. However, RSUs face significant challenges from dynamic workload fluctuations and uneven geographical distribution. These limitations often result in service degradation during peak demand periods. Unmanned Aerial Vehicles (UAVs) offer promising solutions to augment fixed infrastructure capacity. Nevertheless, their energy constraints and trajectory optimization create additional complexity for resource management. To address these challenges, we develop a novel framework integrating workload forecasting with coordinated decision-making for VT migration and UAV routing. We first design a long short-term memory-based workload prediction model. This model predicts workload patterns by combining spatial feature extraction with temporal dependency modeling. We enhance the prediction capability through noise-augmented training to improve robustness. Then, we formulate the VT migration and UAV routing optimization as a markov decision process, which captures the sequential nature of decision-making. Finally, we propose a hierarchical control multi-agent deep reinforcement learning algorithm where the upper-layer controller uses multi-agent proximal policy optimization for collaborative decision-making, and the lower-layer controller handles VT migration and UAV routing execution. Simulation results show that the proposed approach reduces average latency by 25.70% and validation loss by 63.70% for workload prediction compared to baseline methods.

With rapid advancements in large language models (LLMs), AI-generated content (AIGC) has emerged as a key driver of technological innovation and economic transformation. Personalizing AIGC services to meet individual user demands is essential but challenging for AIGC service providers (ASPs) due to the subjective and complex demands of mobile users (MUs), as well as the computational and communication resource constraints faced by ASPs. To tackle these challenges, we first develop a novel multi-dimensional quality-of-experience (QoE) metric. This metric comprehensively evaluates AIGC services by integrating accuracy, token count, and timeliness. We focus on a mobile edge computing (MEC)-enabled AIGC network, consisting of multiple ASPs deploying differentiated AIGC models on edge servers and multiple MUs with heterogeneous QoE requirements requesting AIGC services from ASPs. To incentivize ASPs to provide personalized AIGC services under MEC resource constraints, we propose a QoE-driven incentive mechanism. We formulate the problem as an equilibrium problem with equilibrium constraints (EPEC), where MUs as leaders determine rewards, while ASPs as followers optimize resource allocation. To solve this, we develop a dual-perturbation reward optimization algorithm, reducing the implementation complexity of adaptive pricing. Experimental results demonstrate that our proposed mechanism achieves a reduction of approximately$64.9\%$in average computational and communication overhead, while the average service cost for MUs and the resource consumption of ASPs decrease by$66.5\%$and$76.8\%$, respectively, compared to state-of-the-art benchmarks.

Large Language Models (LLMs) can perform zero-shot learning on unseen tasks and few-shot learning on complex reasoning tasks. However, resource-limited mobile edge networks struggle to support long-context LLM serving for LLM agents during multi-round interactions with users. Unlike stateless computation offloading and static service offloading in edge computing, optimizing LLM serving at edge servers is challenging because LLMs continuously learn from context which raises accuracy, latency, and resource consumption dynamics. In this paper, we propose a joint model caching and inference offloading framework that utilizes test-time deep reinforcement learning (T2DRL) to optimize deployment and execution strategies for long-context LLM serving. In this framework, we analyze the performance convergence and design an optimization problem considering the utilization of context windows in LLMs. Furthermore, the T2DRL algorithm can learn in both the training phase and the testing phase to proactively manage cached models and service requests and adapt to context changes and usage patterns during execution. To further enhance resource allocation efficiency, we propose a double Dutch auction (DDA) mechanism, which dynamically aligns the marginal value of an additional reasoning path with the marginal cost of reasoning services. Finally, experimental results demonstrate that the T2DRL algorithm can reduce system costs by at least 30% compared to baselines while guaranteeing the performance of LLM agents in real-world perception and reasoning tasks.

Edge intelligence in space-air-ground integrated networks (SAGINs) can enable worldwide network coverage beyond geographical limitations for users to access ubiquitous and low-latency intelligence services. Facing global coverage and complex environments in SAGINs, edge intelligence can provision large language models (LLMs) agents for users via edge servers at ground base stations (BSs) or cloud data centers relayed by satellites. As LLMs with billions of parameters are pretrained on vast datasets, LLM agents have few-shot learning capabilities, e.g., chain-of-thought (CoT) prompting for complex tasks, which raises a new trade-off between resource consumption and performance in SAGINs. In this paper, we propose a joint caching and inference framework for edge intelligence to provision sustainable and ubiquitous LLM agents in SAGINs. We introduce “cached model-as-a-resource” for offering LLMs with limited context windows and propose a novel optimization framework, i.e., joint model caching and inference, to utilize cached model resources for provisioning LLM agent services along with communication, computing, and storage resources.We design “age of thought” (AoT) considering the CoT prompting of LLMs, and propose a least AoT cached model replacement algorithm for optimizing the provisioning cost. We propose a deep Q-network-based modified second-bid (DQMSB) auction to incentivize satellite/ground network operators in real-time, which can enhance allocation efficiency by 23% while guaranteeing strategy-proofness and being free from adverse selection.

Rui Pan, Dylan Zhang, Hanning Zhang, Xingyuan Pan, Minrui Xu, Jipeng Zhang, Renjie Pi, Xiaoyu Wang, Tong Zhang. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

Underwater acoustic sensor networks (UASNs) play a key role in ocean resource exploration and complex underwater tasks. However, the open acoustic channel makes them vulnerable to malicious node attacks. Therefore, accurately identifying attack nodes in harsh channels and adapting to their mobility presents a significant challenge. To address these issues, we adopt a meandering ocean current mobility model to describe node movement and construct a hidden Markov model (HMM) along with link transmission loss to characterize the unstable underwater acoustic channel. By monitoring the forwarding behavior of neighboring nodes and combining HMM state inference, we propose a trust model based on a subjective logic framework with dynamic topology updates to detect malicious nodes. It considers variable weights to assess improper node behavior and dynamically updates trustworthiness based on both historical trust and arrival strategies of new and old nodes. Simulation results indicate that the proposed method effectively identifies malicious nodes with attack intensities exceeding 0.38, and for intensities above 0.6, it achieves over 90% identification accuracy and adapts well to dynamic environmental mobility.

Collaborative underwater target hunting, facilitated by multiple autonomous underwater vehicles (AUVs), plays a significant role in various domains, especially military missions. Existing research predominantly focuses on designing efficient and high-success-rate hunting policy, particularly addressing the target's evasion capabilities. However, in real-world scenarios, the target can not only adjust its evasion policy based on its observations and predictions but also possess eavesdropping capabilities. If communication among hunter AUVs, such as hunting policy exchanges, is intercepted by the target, it can adapt its escape policy accordingly, significantly reducing the success rate of the hunting mission. To address this challenge, we propose a covert communication-guaranteed collaborative target hunting framework, which ensures efficient hunting in complex underwater environments while defending against the target's eavesdropping. To the best of our knowledge, this is the first study to incorporate the confidentiality of inter-agent communication into the design of target hunting policy. Furthermore, given the complexity of coordinating multiple AUVs in dynamic and unpredictable environments, we propose an adaptive multi-agent diffusion policy (AMADP), which incorporates the strong generative ability of diffusion models into the multi-agent reinforcement learning (MARL) algorithm. Experimental results demonstrate that AMADP achieves faster convergence and higher hunting success rates while maintaining covertness constraints.

The integration of federated learning (FL) with the Internet of Things (IoT) represents an advanced technological trend, combining the extensive connectivity of IoT with the powerful processing capabilities of FL to drive innovation and optimization across multiple domains. Given the heterogeneity of IoT devices and the variability in data distribution, developing strategies to enhance FL performance without overly burdening resource-constrained devices is crucial. This article proposes an FL algorithm based on generative artificial intelligence (GAI) for IoT devices with extreme heterogeneity in data and resources. The algorithm utilizes pretrained GAI models to generate new data, aligning the data distributions of individual IoT devices closer to independent and identically distributed (i.i.d.), thereby effectively reducing the heterogeneity of local data. Additionally, the proposed algorithm incorporates data synthesis and resource scheduling strategies to mitigate the heterogeneity of local device resources. Finally, we formulate a joint optimization problem aimed at minimizing total energy consumption while maximizing FL performance. Experimental results demonstrate that, under significant resource and data distribution disparities, most existing solutions struggle to converge, whereas the proposed method converges and achieves superior performance. Compared to existing GAI-based approaches, our method significantly reduces latency and energy consumption.

Vehicle Twins (VTs) as digital representations of vehicles can provide users with immersive experiences in vehicular metaverse applications, e.g., Augmented Reality (AR) navigation and embodied intelligence. VT migration is an effective way that migrates the VT when the locations of physical entities keep changing to maintain seamless immersive VT services. However, an efficient VT migration is challenging due to the rapid movement of vehicles, dynamic workloads of Roadside Units (RSUs), and heterogeneous resources of the RSUs. To achieve efficient migration decisions and a minimum latency for the VT migration, we propose a multi-agent split Deep Reinforcement Learning (DRL) framework combined with spatio-temporal trajectory generation. In this framework, multiple split DRL agents utilize split architecture to efficiently determine VT migration decisions. Furthermore, we propose a spatio-temporal trajectory generation algorithm based on trajectory datasets and road network data to simulate vehicle trajectories, enhancing the generalization of the proposed scheme for managing VT migration in dynamic network environments. Finally, experimental results demonstrate that the proposed scheme not only enhances the Quality of Experience (QoE) by 29% but also reduces the computational parameter count by approximately 25% while maintaining similar performances, enhancing users' immersive experiences in vehicular metaverses.

Quantum federated learning (QFL) is an emerging interdisciplinary field that merges the principles of quantum computing (QC) and federated learning (FL), with the goal of leveraging quantum technologies to enhance privacy, security, and efficiency in the learning process. Currently, there is no comprehensive survey for this interdisciplinary field. This review offers a thorough, holistic examination of QFL. We aim to provide a comprehensive understanding of the principles, techniques, and emerging applications of QFL. We discuss the current state of research in this rapidly evolving field, identify challenges and opportunities associated with integrating these technologies, and outline future directions and open research questions. We propose a unique taxonomy of QFL techniques, categorized according to their characteristics and the quantum techniques employed. As the field of QFL continues to progress, we can anticipate further breakthroughs and applications across various industries, driving innovation and addressing challenges related to data privacy, security, and resource optimization. This review serves as a first-of-its-kind comprehensive guide for researchers and practitioners interested in understanding and advancing the field of QFL.

Federated learning (FL) is a promising artificial intelligence framework that enables clients to collectively train models with data privacy. However, in real-world scenarios, to construct practical FL frameworks, several challenges have to be addressed, including statistical heterogeneity, constrained resources, and fairness. Therefore, we first investigate anaggregation gapcaused by statistical heterogeneity during local model initialization, which not only causes additional computational overhead for clients but also leads to the degradation of fairness. To bridge this gap, we proposepFedCal, a novelpersonalizedfederated learning with lightweight adaptivecalibration strategy that performs calibration compensation through the prior knowledge of clients. Specifically, we introduce compensation for each client at the model initialization, with the compensation derived from the global gradient and the latest gradient bias. To enhance the calibration effect, we introduce a smoothing-based calibration strategy, and we design an adaptive calibration strategy. A representative example demonstrates that the proposed calibration and smoothing strategies improve fairness for clients. The theoretical analysis indicates that with an appropriate learning rate, pFedCal converges to a first-order stationary point for non-convex loss functions. Comprehensive experimental results show that pFedCal achieves faster convergence, higher accuracy, and improved fairness than the state-of-the-art methods.

KaShun Shum, Minrui Xu, Jianshu Zhang, Zixin Chen, Shizhe Diao, Hanze Dong, Jipeng Zhang, Muhammad Omer Raza. Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing. 2024.

Vehicle trajectory data plays a pivotal role in simulation testing for autonomous driving. Hence, there exist well-established trajectory generation methods employing deep generative models to generate trajectories mapping the distribution of the original dataset, thereby augmenting existing trajectory datasets. However, these methods typically rely on large datasets gathered by governmental or organizational entities for central training, which may pose data privacy, security, and accessibility issues. Therefore, it is challenging to generate high-quality traffic trajectory data while preserving privacy which involves a delicate balance between these two objectives. To deal with this challenge, we introduce Federated Learning into the diffusion model and propose a Federated Learning-based diffusion model (FedDifftraj) to generate traffic trajectory data. Unlike existing central training methods, FedDifftraj aggregates model parameters uploaded by different vehicles and then updates a global model. Additionally, there is a substantial communication overhead incurred during the training of the federated diffusion model. Therefore, we quantize the local diffusion model before uploading it to the parameter server. Through extensive simulations on real-world datasets, FedDifftraj can generate high-quality traffic trajectory data that is consistent with the results of the central training while preserving privacy and reducing communication overhead by 93.74% when utilizing 8-bit quantization.

Air-ground integrated networks can relieve communication pressure on ground transportation networks and provide 6G-enabled vehicular Metaverses services offloading in remote areas with sparse RoadSide Units (RSUs) coverage and downtown areas where users have a high demand for vehicular services. Vehicle Twins (VTs) are the digital twins of physical vehicles to enable more immersive and realistic vehicular services, which can be offloaded and updated on RSU, to manage and provide vehicular Metaverses services to passengers and drivers. The high mobility of vehicles and the limited coverage of RSU signals necessitate VT migration to ensure service continuity when vehicles leave the signal coverage of RSUs. However, uneven VT task migration might overload some RSUs, which might result in increased service latency, and thus impactive immersive experiences for users. In this paper, we propose a dynamic Unmanned Aerial Vehicle (UAV)-assisted VT migration framework in air-ground integrated networks, where UAVs act as aerial edge servers to assist ground RSUs during VT task offloading. In this framework, we propose a diffusion-based Reinforcement Learning (RL) algorithm, which can efficiently make immersive VT migration decisions in UAV-assisted vehicular networks. To balance the workload of RSUs and improve VT migration quality, we design a novel dynamic path planning algorithm based on a heuristic search strategy for UAVs. Simulation results show that the diffusion-based RL algorithm with UAV-assisted performs better than other baseline schemes.

With the advances of Virtual Reality (VR) / Augmented Reality (AR), there arises a compelling need for transmission of point clouds over lossy channels (e.g., a 5G millimeter wave (mmWave) link that tends to be easily blocked). In this paper, we revisit the traditional Multiple Description Coding (MDC) concept and propose a simple point cloud MDC scheme that takes advantage of voxelization and is built upon a typical geometric point cloud compression codec. Our simulation study demonstrates the efficacy of the proposed scheme, as well as the tradeoff between compression efficiency and point cloud quality gain offered by MDC.

Generative Artificial Intelligence (GAI) shows remarkable productivity and creativity in Mobile Edge Networks, such as the metaverse and the Industrial Internet of Things. Federated learning is a promising technique for effectively training GAI models in mobile edge networks due to its data distribution. However, there is a notable issue with communication consumption when training large GAI models like generative diffusion models in mobile edge networks. Additionally, the substantial energy consumption associated with training diffusion-based models, along with the limited resources of edge devices and complexities of network environments, pose challenges for improving the training efficiency of GAI models. To address this challenge, we propose an on-demand quantized energy-efficient federated diffusion approach for mobile edge networks. Specifically, we first design a dynamic quantized federated diffusion training scheme considering various demands from the edge devices. Then, we study an energy efficiency problem based on specific quantization requirements. Numerical results show that our proposed method significantly reduces system energy consumption and transmitted model size compared to both baseline federated diffusion and fixed quantized federated diffusion methods while effectively maintaining reasonable quality and diversity of generated data.

The development of LEO satellite networks (LSN) makes them a potential solution to deliver broadcast/multicast traffic to deploy and upgrade massive amounts of Internet of Things (IoT) devices in future 6G networks. However, inherent resource constraints of LSN leave them vulnerable to a multitude of security threats, most notably distributed denial-of-service (DDoS) attacks. Existing solutions are primarily based on machine learning detection methods which are incapable of defending against unknown zero-day attacks. This paper presents an innovative solution leveraging deep reinforcement learning (DRL) to create a dynamic multicast tree based on moving target defense (MTD), aimed at enhancing the security of multicast services in LSN. The proposed solution adopts an adaptive orbital tree mutation (AOTM) scheme that dynamically adjusts multicast tree configurations considering quality of service (QoS) constraints to avoid attacks on vulnerable nodes. Simulations demonstrate the effectiveness of the AOTM scheme, showcasing its superior defense success rates compared to existing state-of-the-art algorithms.

This paper proposes a hybrid Quantum Federated Learning (QFL) method, called QQFL, a revolutionary approach for Dynamic Security Assessment (DSA) optimized for modern smart grids. Built on the synergy of measurement-device-independent QKD (MDI-QKD) and Variational Quantum Circuit (VQC), QQFL uniquely addresses the challenges of centralized structures and vulnerabilities in existing ML-based DSA techniques. It enables accurate label predictions for quantum states encoded from classical DSA data while ensuring data security via QKD networks. A novel mechanism, the DNN-based MDI-QKD optimizer, ensures optimal secret key exchange. Unlike traditional methods reliant solely on classical CPUs, QQFL integrates QPUs for executing computational tasks. Given the imperative of frequent data transmission in modern rapidly changing smart grid environment, QQFL emphasizes swift online learning and dynamic deployment. Testing on the synthetic Illinois 49-machine 200-bus system affirms QQFL's superior the DSA accuracy while upholding the data privacy of smart grids. Ultimately, QQFL enhances the security, reliability, confidentiality, and robustness of sophisticated smart grids.

Transforming the multi-round vanilla Federated Learning (FL) into one-shot FL (OFL) significantly reduces the communication burden and makes a big leap toward practical deployment. However, we note that existing OFL methods all build on model lossy reconstruction (i.e., aggregating while partially discarding local knowledge in clients' models), which attains one-shot at the cost of degraded inference performance. By identifying the root cause of stressing too much on finding a one-fit-all model, this work proposes a novel one-shot FL framework by embodying each local model as an independent expert and leveraging a Mixture-of-Experts network to maintain all local knowledge intact. A dedicated self-supervised training process is designed to tune the network, where the sample generation is guided by approximating underlying distributions of local data and making distinct predictions among experts. Notably, the framework also fuels FL with flexible, data-free aggregation and heterogeneity tolerance. Experiments on 4 datasets show that the proposed framework maintains the one-shot efficiency, facilitates superior performance compared with 8 OFL baselines (+5.54% on CIFAR-10), and even attains over ×4 performance gain compared with 3 multi-round FL methods, while only requiring less than 85% trainable parameters. Our code will be available at https://github.com/zenghui9977/IntactOFL.

Interdisciplinary studies often require researchers to explore literature in diverse branches of knowledge. Yet, navigating through the highly scattered knowledge from unfamiliar disciplines poses a significant challenge. In this paper, we introduce DiscipLink, a novel interactive system that facilitates collaboration between researchers and large language models (LLMs) in interdisciplinary information seeking (IIS). Based on users’ topic of interest, DiscipLink initiates exploratory questions from the perspectives of possible relevant fields of study, and users can further tailor these questions. DiscipLink then supports users in searching and screening papers under selected questions by automatically expanding queries with disciplinary-specific terminologies, extracting themes from retrieved papers, and highlighting the connections between papers and questions. Our evaluation, comprising a within-subject comparative experiment and an open-ended exploratory study, reveals that DiscipLink can effectively support researchers in breaking down disciplinary boundaries and integrating scattered knowledge in diverse fields. The findings underscore the potential of LLM-powered tools in fostering information-seeking practices and bolstering interdisciplinary research.

Enhanced by machine learning (ML) techniques, data-driven dynamic security assessment (DSA) in smart cyber-physical grids has attracted great research interests in recent years. However, as existing DSA methods generally rely on centralized ML architectures, the scalability, privacy, and cost effectiveness of existing methods are limited. To address these issues, we propose a novel quantum-secured distributed intelligent system for smart cyber-physical DSA based on Federated learning (FL) and quantum key distribution (QKD), namely, quantum-secured federated DSA (QFDSA). QFDSA aggregates the knowledge learned from various local data owners (also known as clients) to predict and evaluate the system stability status in a decentralized fashion. In addition, in order to preserve the privacy of the distributed DSA data, QFDSA adopts the measurement-device-independent QKD, which can further improve the security of local DSA model transmission. Moreover, to accommodate the typical fast system environment and requirement changes, QFDSA alleviates the issues of limited key generation rates by utilizing secret-key pool that guarantee the availability of adequate secret-key materials. Extensive experiments based on the New England 10-machine 39-bus testing system and the synthetic Illinois 49-machine 200-bus testing system demonstrate that the proposed QFDSA method can achieve more advantageous DSA performance while protecting the privacy of local data for real-time DSA applications compared to the benchmarks. Besides, the secret-key generation rate can be improved to adjust its parameters dynamically in real time.

Avatars, as promising digital assistants in Vehicular Metaverses, can enable drivers and passengers to immerse in 3-D virtual spaces, serving as a practical emerging example of Artificial Intelligence of Things (AIoT) in intelligent vehicular environments. The immersive experience is achieved through seamless human–avatar interaction, e.g., augmented reality navigation, which requires intensive resources that are inefficient and impractical to process on intelligent vehicles locally. Fortunately, offloading avatar tasks to roadside units (RSUs) or cloud servers for remote execution can effectively reduce resource consumption. However, the high mobility of vehicles, the dynamic workload of RSUs, and the heterogeneity of RSUs pose novel challenges to making avatar migration decisions. To address these challenges, in this article, we propose a dynamic migration framework for avatar tasks based on real-time trajectory prediction and multiagent deep reinforcement learning (MADRL). Specifically, we propose a model to predict the future trajectories of intelligent vehicles based on their historical data, indicating the future workloads of RSUs. Based on the expected workloads of RSUs, we formulate the avatar task migration problem as a long-term mixed-integer programming problem. To tackle this problem efficiently, the problem is transformed into a partially observable Markov decision process (POMDP) and solved by multiple DRL agents with hybrid continuous and discrete actions in decentralized. Numerical results demonstrate that our proposed algorithm can effectively reduce the latency of executing avatar tasks by around 25% without prediction and 30% with prediction and enhance user immersive experiences in the AIoT-enabled Vehicular Metaverse (AeVeM).

The increasing focus on privacy and security in 6G networks, which are intelligence-native, necessitates the use of quantum key distribution-secured semantic information communication (QKD-SIC) to protect confidential data. In QKD-SIC systems, edge devices connected via quantum channels can efficiently encrypt semantic information from the semantic source, and securely transmit the encrypted semantic information to the semantic destination. In this article, we consider an efficient resource (i.e., quantum key distribution (QKD) and KM wavelengths) sharing problem to support QKD-SIC systems under the uncertainty of semantic information generated by edge devices. In such a system, QKD service providers offer QKD services with different subscription options to the edge devices. The QKD services are envisioned to follow cloud computing that has the subscription in the reservation and on-demand options, i.e., for long and short (immediate) terms, respectively. As such, to reduce the cost for the edge device users, we propose a QKD resource management framework for the edge devices communicating semantic information. The framework is based on a two-stage stochastic optimization model to achieve optimal QKD deployment. Moreover, to reduce the deployment cost of QKD service providers, QKD resources in the proposed framework can be utilized based on efficient QKD-SIC resource management, including semantic information transmission among edge devices, secret-key provisioning, and cooperation formation among QKD service providers. In detail, the formulated two-stage stochastic optimization model can achieve the optimal QKD-SIC resource deployment while meeting the secret-key requirements for semantic information transmission of edge devices. Moreover, to share the cost of the QKD resource pool among cooperative QKD service providers forming a coalition in a fair and interpretable manner, the proposed framework leverages the concept of Shapley value from cooperative game theory as a solution. Experimental results demonstrate that the proposed framework can reduce the deployment cost by about 40% compared with existing noncooperative baselines.

The synergy between Unmanned Aerial Vehicles (UAVs) and metaverses is giving rise to an emerging paradigm named UAV metaverses, which create a unified ecosystem that blends physical and virtual spaces, transforming drone interaction and virtual exploration. UAV Twins (UTs), as the digital twins of UAVs that revolutionize UAV applications by making them more immersive, realistic, and informative, are deployed and updated on ground base stations, e.g., RoadSide Units (RSUs), to offer metaverse services for UAV Metaverse Users (UMUs). Due to the dynamic mobility of UAVs and limited communication coverages of RSUs, it is essential to perform real-time UT migration to ensure seamless immersive experiences for UMUs. However, selecting appropriate RSUs and optimizing the required bandwidth is challenging for achieving reliable and efficient UT migration. To address the challenges, we propose a tiny machine learning-based Stackelberg game framework based on pruning techniques for efficient UT migration in UAV metaverses. Specifically, we formulate a multi-leader multifollower Stackelberg model considering a new immersion metric of UMUs in the utilities of UAVs. Then, we design a Tiny Multi-Agent Deep Reinforcement Learning (Tiny MADRL) algorithm to obtain the tiny networks representing the optimal game solution. Specifically, the actor-critic network leverages the pruning techniques to reduce the number of network parameters and achieve model size and computation reduction, allowing for efficient implementation of Tiny MADRL. Numerical results demonstrate that our proposed schemes have better performance than traditional schemes.

5G and Beyond 5G communication technologies have promoted the architectural innovation of the Industrial Internet of Things (IIoT) and the wide application of edge learning. As Beyond 5G technologies enhance wireless communication within IIoT, the demand for efficient, secure data management becomes paramount. Edge learning emerges as a solution for localized model training, reducing the necessity for extensive data transmission. However, this decentralization introduces vulnerabilities, particularly in data security during transmission and efficient resource utilization. To address the challenges of data scheduling for edge learning in the Wireless IIoT (WIIoT), we propose a novel architecture that leverages blockchain for secure, decentralized data scheduling and employs physically unclonable functions (PUFs)-based algorithm to ensure data integrity and confidentiality. The primary contributions consist of a task scheduling model based on blockchain, along with a data compression scheme in multiple stages combined with a data scheduling algorithm that is optimized for energy efficiency in edge learning environments. Experiments conducted on a simulated WIIoT platform comprising embedded devices validate our approach, demonstrating enhanced data security and learning efficiency which can reduce 40% in the training stage and 70% in the inference stage. Our findings contribute to the advancement of security and efficient edge learning frameworks in the context of WIIoT, addressing the intricate balance between security, efficiency, and decentralized trust.

The integration of edge computing with digital twins (DTs) has been instrumental in driving substantial advancements in the Internet of Vehicles (IoV) domain in recent times, particularly within the 6G wireless networks where DTs enable real-time simulation, monitoring, analysis, and high-speed transmissions for connected vehicles. Despite these benefits, several challenges arise, including dynamic network topologies resulting from the high-speed vehicle mobility, frequent edge server switches causing instability and increased latency, and the limited computing resources struggling to cope with the demanding computational tasks. This article addresses these issues by proposing a framework where the vehicles serve as the auxiliary mobile edge computing (MEC) servers. It introduces an enhanced density-based spatial clustering of applications with the noise (DBSCAN) algorithm designed to improve the clustering of vehicles under high-speed movement scenarios. Moreover, a multi-to-multi matching algorithm is devised to effectively associate vehicles with the auxiliary MEC servers. To alleviate the problem of insufficient computing resources due to intense computational loads during DT updates, a deep reinforcement learning (DRL)-based approach is utilized to make the optimal computation offloading decisions. This work further refines the offloading strategy by adopting the improved double deep Q-network (DDQN) and the dueling deep Q-network algorithms. Simulation experiments validate that the proposed clustering improvement and the DRL-based offloading decision-making scheme outperform the existing baseline methods across multiple performance metrics, such as clustering effectiveness, processing latency reduction, algorithmic efficiency, and convergence rate.

Metaverse can obscure the boundary between the physical and virtual worlds. Specifically, for the human-centric Metaverse in vehicular networks, i.e., the vehicular Metaverse, vehicles are no longer isolated physical spaces but interfaces that extend the virtual worlds to the physical world. Accessing the human-centric Metaverse via autonomous vehicles (AVs), drivers and passengers can immerse in and interact with 3D virtual objects overlaying views of streets on head-up displays (HUD) via augmented reality (AR). The seamless, immersive, and interactive experience rather relies on real-time multi-dimensional data synchronization between physical entities, i.e., AVs, and virtual entities, i.e., Metaverse billboard providers (MBPs). However, mechanisms to allocate and match synchronizing AV and MBP pairs to roadside units (RSUs) in a synchronization service market, which consists of the physical and virtual submarkets, are vulnerable to adverse selection. In this paper, we propose an enhanced second-score auction-based mechanism, named EPViSA, to allocate physical and virtual entities in the synchronization service market of the vehicular Metaverse. The EPViSA mechanism can determine synchronizing AV and MBP pairs simultaneously while protecting participants from adverse selection and thus achieving high total social welfare. We propose a synchronization scoring rule to eliminate the external effects from the virtual submarkets. Then, a price scaling factor is introduced to enhance the allocation of synchronizing virtual entities in the virtual submarkets. Finally, rigorous analysis and extensive experiments demonstrate EPViSA can achieve at least 96% of the social welfare compared to the omniscient benchmark while ensuring strategy-proof and adverse selection free through a simulation testbed.

Web 3.0 pursues the establishment of decentralized ecosystems through blockchain technologies, driving digital transformation in commerce and governance. With consensus algorithms and smart contracts grounded in cryptographic technologies, Web 3.0 enables secure and transparent digital services, such as digital identity, asset management, decentralized autonomous organizations (DAOs), and decentralized finance (DeFi), fostering integration between digital and physical economies. As quantum devices rapidly advance, Web 3.0 is being developed in parallel with the deployment of quantum cloud computing and quantum Internet. In this regard, quantum computing first disrupts the original cryptographic systems that protect data security while reshaping modern cryptography with enhanced quantum computing and communication capabilities. This article provides a comprehensive overview of blockchain-based Web 3.0, examining its quantum and postquantum advancements from two key perspectives. On the one hand, postquantum migration methods and quantum-resistant signatures offer robust solutions to safeguard blockchain against quantum threats. On the other hand, quantum and postquantum encryption and verification algorithms boost blockchain performance, creating a decentralized, secure, and value-driven system. Additionally, we outline potential applications of quantum blockchain and offer guidance for implementation within the Web 3.0 ecosystem. Finally, we discuss future directions for developing a provably secure and decentralized digital ecosystem.

Due to the scarcity of spectrum resources in Non-orthogonal Multiple Access (NOMA) systems and insufficient satellite-ground integration in satellite Internet of Things (IoT), this paper investigates its issue in spectrum resource management. We propose a resource allocation method based on Multi-Agent Deep Deterministic Policy Gradient (MADDPG) for NOMA enabled satellite IoT. We formulate the spectrum allocation problem of the satellite-ground integrated network as a distributed optimization problem. Then we decouple the problem into two sub-problems. Firstly, a user grouping method based on matching coefficients is defined, and a Linear Programming (LP) method is utilized for obtaining solution. Secondly, the power allocation problem is transformed into a multi-agent problem, where MADDPG is employed to allocate the power. Through this approach, the system is capable of real-time user association and spectrum resource allocation optimization, achieving optimal user grouping while maximizing system transmission rate. Based on the simulation results, the MADDPG-based method demonstrates fast convergence within 100 training iterations. The proposed MADDPG-based resource management method also achieves increased system transmission rate with more effective matching outcomes over Deep Deterministic Policy Gradient (DDPG), Orthogonal Multiple Access (OMA), and random allocation baselines.

Mobile edge computing (MEC) is a key enabling technology for the sixth-generation (6G) wireless networks. In this paper, we apply covert communications to MEC to prevent information leakage, where two candidate technologies of 6G, reconfigurable intelligent surface (RIS) and non-orthogonal multiple access (NOMA), are adopted. Specifically, a legitimate transmitter sends messages to a pair of legitimate receivers, while a warden aims to detect whether the legitimate transmission exists. We can hide the existence of the stronger-signal receiver's transmission from the warden by exploiting the nature of NOMA, and we use a jammer to further hide this existence. We first analyze the performance for the case of fixed power allocation between the legitimate transmitters and the jammer. The closed-form expressions for the minimum detection error probability and ergodic public/covert rates are derived. Then, we design a reinforcement learning (RL)-based power-allocation optimization algorithm that maximizes the sum rate while ensuring covertness, by optimizing the power allocation between the transmitters and the jammer. Simulation results validate the correctness of our analysis and demonstrate the covertness of the proposed scheme. Furthermore, the performance of the RL-based algorithm is significantly better than that of the baseline scheme, which reflects the effectiveness of our proposed algorithm.

Digital Twin (DT) technologies create digital models of physical entities frequently in multimedia forms, which are crucial for concurrent simulation and analysis of real-world systems. In displaying DTs, Holographic-Type Communication (HTC) provides immersive multimedia access for users to interact with Holographic DTs (HDTs) by transmitting holographic data such as Light Field (LF) and other multisensory information. HDT has applications in remote education, work, and social interactions. However, the effective matching of demand and supply between HDT users and providers remains a challenge. To address this issue, we propose a hierarchical architecture that integrates the DT and HTC paradigms. This architecture incorporates a marketplace for HDT services, leveraging a formulated Double Dutch Auction (DDA) mechanism to optimize matching and pricing based on user and provider valuation. Furthermore, We employ an actor-critic-based Deep Reinforcement Learning (DRL) algorithm to train a DDA auctioneer that dynamically adjusts auction clocks during the auction process. As an alternative to the Multi-layer Perceptron (MLP), we experiment with a Deep Simplistic Variational Quantum Circuit (DSVQC) to reduce the number of parameters and enhance performance stability. Our simulations reveal that the proposed learning-based auctioneer achieves 92% optimal social welfare at a 37% auction information exchange cost for an MLP-based actor and 99% optimal social welfare at a 77% auction information exchange cost for a DSVQC-based actor.

Edge computing has evolved to enable mobile applications to run in an efficient and cost-effective manner at explosive-growing edge nodes. Under this paradigm, a new business resource trading market has emerged to provide edge-cloud services, offering a convenient way for mobile users to obtain resources from distributed computing power providers (CPPs). Blockchain, as a promising technology, provides a reliable platform for multi-party resource transactions (TXs), enabling secure and reliable computing services. Notably, the distributed CPPs not only offer mobile services but also act as blockchain nodes to maintain the stability of TXs. In this case, there exist certain bottlenecks in the blockchain-enabled edge-cloud resource market, such as limited scalability, inefficient resource allocation, and large system cost. In this paper, assisted by the permissioned blockchain, we study the fundamental problem of resource allocation by minimizing the system cost to handle mobile services and blockchain TXs in parallel. We first partition the Practical Byzantine Fault Tolerant (PBFT) consensus by hierarchical sharding to improve the scalability and ensure the security of the blockchain system. Next, based on the optimal sharding strategies, we formulate the parallel resource allocation as a multi-scale Lyapunov optimization problem, and develop a dual-alternation actor-critic with an attention mechanism (DA3C) algorithm to solve it. We evaluate the performance of theParamartusing trace-driven experiments. Simulation results demonstrate the superiority of our proposed framework as compared with the benchmark algorithms.

With the rapid development of artificial general intelligence (AGI), various multimedia services based on pretrained foundation models (PFMs) need to be effectively deployed. With edge servers that have cloud-level computing power, edge intelligence can extend the capabilities of AGI to mobile edge networks. However, compared with cloud data centers, resource-limited edge servers can only cache and execute a small number of PFMs, which typically consist of billions of parameters and require intensive computing power and GPU memory during inference. To address this challenge, in this paper, we propose a joint foundation model caching and inference framework that aims to balance the tradeoff among inference latency, accuracy, and resource consumption by managing cached PFMs and user requests efficiently during the provisioning of generative AI services. Specifically, considering the in-context learning ability of PFMs, a new metric named the Age of Context (AoC), is proposed to model the freshness and relevance between examples in past demonstrations and current service requests. Based on the AoC, we propose a least context caching algorithm to manage cached PFMs at edge servers with historical prompts and inference results. The numerical results demonstrate that the proposed algorithm can reduce system costs compared with existing baselines by effectively utilizing contextual information.

Recent breakthroughs in generative artificial intelligence have triggered a surge in demand for machine learning training, which poses significant cost burdens and environmental challenges due to its substantial energy consumption. Scheduling training jobs among geographically distributed cloud data centers unveils the opportunity to optimize the usage of computing capacity powered by inexpensive and low-carbon energy and address the issue of workload imbalance. To tackle the challenge of multi-objective scheduling, i.e., maximizing GPU utilization while reducing operational costs, we propose an algorithm based on multi-agent reinforcement learning and actor-critic methods to learn the optimal collaborative scheduling strategy through interacting with a cloud system built with real-life workload patterns, energy prices, and carbon intensities. Compared with other algorithms, our proposed method improves the system utility by up to 28.6% attributable to higher GPU utilization, lower energy cost, and less carbon emission.

In this paper, a novel paradigm of mobile edgequantum computing (MEQC) is proposed, which brings quantum computing capacities to mobile edge networks that are closer to mobile users (i.e., edge devices). First, we propose an MEQC system model where mobile users can offload computational tasks to scalable quantum computers via edge servers with cryogenic components and fault-tolerant schemes. Second, we show that it is NP-hard to obtain a centralized solution to the partial offloading problem in MEQC in terms of the optimal latency and energy cost of classical and quantum computing. Third, we propose a multi-agent hybrid discrete-continuous deep reinforcement learning using proximal policy optimization to learn the long-term sustainable offloading strategy without prior knowledge. Finally, experimental results demonstrate that the proposed algorithm can reduce at least 30% of the cost compared with the existing baseline solutions under different system settings.

Quantum cloud computing is a promising paradigm for efficiently provisioning quantum resources (i.e., qubits) to users. In quantum cloud computing, quantum cloud providers provision quantum resources in reservation and on-demand plans for users. Literally, the cost of quantum resources in the reservation plan is expected to be cheaper than the cost of quantum resources in the on-demand plan. However, quantum resources in the reservation plan have to be reserved in advance without information about the requirement of quantum circuits beforehand, and consequently, the resources are insufficient, i.e., under-reservation. Hence, quantum resources in the on-demand plan can be used to compensate for the unsatisfied quantum resources required. To end this, we propose a quantum resource allocation for the quantum cloud computing system in which quantum resources and the minimum waiting time of quantum circuits are jointly optimized. Particularly, the objective is to minimize the total costs of quantum circuits under uncertainties regarding qubit requirement and minimum waiting time of quantum circuits. In experiments, practical circuits of quantum Fourier transform are applied to evaluate the proposed qubit resource allocation. The results illustrate that the proposed qubit resource allocation can achieve the optimal total costs.

Artificial Intelligence-Generated Content (AIGC) refers to the paradigm of automated content generation utilizing AI models. Mobile AIGC services in the Internet of Vehicles (IoV) network have numerous advantages over traditional cloud-based AIGC services, including enhanced network efficiency, better reconfigurability, and stronger data security and privacy. Nonetheless, AIGC service provisioning frequently demands significant resources. Consequently, resource-constrained roadside units (RSUs) face challenges in maintaining a heterogeneous pool of AIGC services and addressing all user service requests without degrading overall performance. Therefore, in this paper, we propose a decentralized incentive mechanism for mobile AIGC service allocation, employing multi-agent deep reinforcement learning to find the balance between the supply of AIGC services on RSUs and user demand for services within the IoV context, optimizing user experience and minimizing transmission latency. Experimental results demonstrate that our approach achieves superior performance compared to other baseline models.

As healthcare becomes increasingly data-driven, integrating hybrid mobile edge-quantum computing (MEQC) into smart healthcare systems emerges as a promising solution for handling growing computational demand, especially for latency-sensitive tasks. Therefore, this paper proposes a deep reinforcement learning (DRL)-based Lyapunov approach for schedule computation offloading, aiming to minimize the total latency in hybrid MEQC-based smart healthcare systems. In this framework, a sustainable computation offloading strategy is obtained while guaranteeing the individual latency constraints and the required success ratio for each computation task. More precisely, the original latency minimization problem is transformed into a stepwise mixed-integer non-convex optimization problem using Lyapunov techniques. Subsequently, a Deep Q-Network (DQN) is adopted for computation offloading mode selection. The effectiveness of the proposed approach and its dependency on various system parameters are validated and assessed through numerical simulations.

Fastest route recommendation (FRR) is crucial for intelligent transportation systems. The existing methods treat it as a pathfinding problem on dynamic graphs, and extend A* algorithm with neuralized travel time estimators as cost functions. However, they fail to provide effective heuristic cost due to the neglect of its admissibility and the utilization of noise path information, resulting in sub-optimal results and inefficiency. Besides, path sequentiality is also ignored, affecting algorithm accuracy as well. In this paper, we propose a variational inference based fastest route recommendation method, which follows the framework of A* algorithm and provides effective costs for routing. Specifically, we first adopt a sequential estimator to accurately estimate the travel time of a specific path. More importantly, we design a variational inference based estimator, which models the distribution of travel time between two nodes and provides an effective heuristic cost with high probability of being admissible. We further take advantage of adversarial learning to enrich the fastest path information. To the best of our knowledge, we are the first to use variational estimator to consider the admissibility of heuristics in FRR. Extensive experiments are conducted on two real-world datasets. The results verify the performance advantage of our proposed method.

In this paper, we apply covert communications to reconfigurable intelligent surface (RIS)-assisted non-orthogonal multiple access (NOMA) networks, where a legitimate transmitter sends messages to a pair of legitimate users while a warden aims to detect whether the legitimate transmission exists. We can hide the existence of the strong user's transmission from a warden by exploiting the nature of NOMA, i.e., allocating less power to the strong user, and we use a jammer to further hide that existence. Correspondingly, we analyze the system performance and obtain the closed-form expression for the minimum detection error probability. Simulation results validate the correctness of our analysis and demonstrate the covertness of the proposed scheme.

Federated learning (FL) is a distributed machine learning paradigm with a promising future, which can preserve data privacy while training the global model collaboratively. However, FL is still facing model confidentiality issues. Therefore, in this paper, we propose a quantum key distribution (QKD) based secure FL scheme to facilitate FL model encryption against network eavesdropping attacks. Specifically, we introduce a stochastic resource allocation scheme for QKD to support FL networks. In the network, remote FL workers are connected to the server to train an aggregated global model in a distributed manner. However, due to the unpredictable number of workers at each location, the demand for secret-key rates to support secure model transmission to the server is not uniform. The proposed scheme can allocate QKD resources (i.e., wavelengths) in a way that minimizes the total cost given the stochastic demand. We formulate the optimization problem for the proposed scheme as a stochastic programming model. Numerical results demonstrate that the proposed scheme can successfully achieve the cost-minimizing objective while satisfying all uncertain demands and other security constraints.

The Metaverse is regarded as the next-generation Internet paradigm that allows humans to play, work, and socialize in an alternative virtual world with an immersive experience, for instance, via head-mounted displays for Virtual Reality (VR) rendering. With the help of ubiquitous wireless connections and powerful edge computing technologies, VR users in the wireless edge-empowered Metaverse can immerse themselves in the virtual through the access of VR services offered by different providers. However, VR applications are computation- and communication-intensive. The VR service providers (SPs) have to optimize the VR service delivery efficiently and economically given their limited communication and computation resources. An incentive mechanism can be thus applied as an effective tool for managing VR services between providers and users. Therefore, in this paper, we propose a learning-based Incentive Mechanism framework for VR services in the Metaverse. First, we propose the quality of perceptual experience as the metric for VR users immersing in the virtual world. Second, for quick trading of VR services between VR users (i.e., buyers) and VR SPs (i.e., sellers), we design a double Dutch auction mechanism to determine optimal pricing and allocation rules in this market. Third, for auction information exchange cost reduction, we design a deep reinforcement learning-based auctioneer to accelerate this auction process. Experimental results demonstrate that the proposed framework can achieve near-optimal social welfare while reducing at least half of the auction information exchange cost than baseline methods.

Federated learning (FL) is an emerging technology for empowering various applications that generate large amounts of data in intelligent cyber–physical systems (ICPS). Though FL can address users’ concerns about data privacy, its maintenance still depends on efficient incentive mechanisms. For long-term incentivization to participants in data federation under dynamic environments, deep reinforcement learning as a promising technology has been extensively studied. However, the nonstationary problem caused by the heterogeneity of ICPS devices results in a serious effect on the convergence rate of existing single-agent reinforcement learning. In this article, we propose a multiagent learning-based incentive mechanism to capture the stationarity approximation in FL with heterogeneous ICPS. First, we formulate the secure communication and data resource allocation problem as a Stackelberg game in FL with multiple participants. Then, to tackle the heterogeneous problem, we model this multiagent game as a partially observable Markov decision process. In particular, a multiagent federated reinforcement learning algorithm is proposed to learn the allocation policies efficiently by dwindling variances in policy evaluation caused by interaction among multiple devices without the requirement of sharing privacy information. Moreover, the proposed algorithm is proved to attain convergence at an expected rate. Finally, extensive experimental results demonstrate that our proposed algorithm significantly outperforms baseline approaches.

Dynamic off-chain routing in payment channel network (PCN)-based Internet of Things (IoT) is attracting increasing research attention. However, there are two major issues in dynamic routing in PCN-based IoT with resource-limited devices. The first issue is how to achieve high long-term transaction efficiency in PCN with dynamic channel capacities. The second issue is how to achieve a lightweight routing algorithm deployed on IoT devices while achieving high transaction efficiency, i.e., successful payment amount and success ratio. Therefore, in this paper, we propose a compact deep reinforcement learning (DRL) algorithm to learn the joint dynamic and lightweight routing policy for maximizing long-term transaction efficiency. To obtain optimal performance in dynamic routing problems for off-chain systems, a proximal policy optimization algorithm is employed to create an actor–critic learning structure for training the teacher DRL model. To obtain a compact and efficient student DRL model, an adaptive pruning technique is utilized for pruning unnecessary parameters of networks in the teacher model adaptively without affecting its learning ability. Furthermore, knowledge distillation is leveraged to improve the performance of the student network. Thus, a compact and efficient student DRL model can be developed and implemented to maximize the long-term transaction efficiency in off-chain systems on resource-limited IoT devices. The simulation results demonstrate that the proposed DRL algorithm outperforms the other baseline algorithms in PCN transaction efficiency while requiring only 10% of the computation and storage resources compared with that of the original teacher model.

Blockchain technologies allow the Internet of Things (IoT) to build trust among various interest parties. For the resource-limited IoT devices, offloading computation-intensive tasks (blockchain verification and mining tasks, and data process tasks) to edge servers for execution is considered as a promising solution in mobile-edge computing. However, conventional methods (such as linear programming or game theory) for the computation offloading problem cannot achieve long-term performance while the existing deep reinforcement learning (DRL)-based algorithms suffer from slow convergence, lack of robustness, and unstable performance. In this article, we propose a multiagent DRL framework to achieve long-term performance for cooperative computation offloading, in which a scatter network is adopted to improve its stability and league learning is introduced for agents to explore the environment collaboratively for fast convergence and robustness. First, we study the nonorthogonal multiple access-enabled cooperative computation offloading problem and formulate the joint problem as a Markov decision process by considering both the blockchain mining tasks and data processing tasks. Second, to avoid useless exploration and unstable performance, we initially train an intelligent agent represented by scatter networks using conventional expert strategies. Third, in order to enhance the performance, we subsequently establish a hierarchical league where agents collaborate with others to explore the environment. Finally, our experimental results demonstrate that our algorithm could perform better in terms of reducing energy cost and delay cost, and shortening almost 60% of the training time compared with the state-of-the-art approaches.