The growing demand for intelligent mobile applications has made the deployment and operation of Deep Neural Networks (DNNs) on mobile Edge Devices (EDs) increasingly essential. However, the highly dynamic nature of edge environments and the limited computational resources of EDs result in significant energy consumption and compromised inference quality. To address these issues, we propose the Demand-Aware Evolutionary Reinforcement Learning (DA-ERL) framework, a novel approach for optimizing Partitioned Collaborative Inference (PCI) across multiple EDs and Mobile Edge Computing (MEC) servers. At the core of DA-ERL is a Demand-Aware Spatio-Temporal Graph Convolutional Network (DA-STGCN). This new architecture creates a predictive state representation by uniquely integrating two channels: a Spatial Graph Channel using Graph Convolutional Networks to model the network topology, and a Temporal Prediction Channel using Temporal Convolutional Networks to capture the evolution of system dynamics. Moreover, we design and formulate a task dynamic demand index to model the dynamic task characteristics, which guides the agent's learning policy. Furthermore, we train DA-ERL within a Cross-Entropy Method (CEM) based evolutionary framework that leverages elite-guided exploration to enhance sample efficiency in complex search spaces. Extensive simulations demonstrate that the proposed DA-ERL framework significantly outperforms conventional methods, achieving a 23.4% reduction in system cost while maintaining a near-perfect task completion rate in high-density scenarios.

Low Earth Orbit (LEO) satellite networks hold great promise in the field of wireless communication due to their global coverage. However, the long communication distances and massive data computations present significant challenges for current satellite networks. To overcome these barriers, we propose SkyLink, a universal Integrated Ground-Air-Space Collaborative Edge Computing system that leverages horizontal collaboration among aerial platforms (AirXs) as well as vertical collaboration among Ground-Air-Space. We propose a bi-level optimization framework based on a Multi-Agent Twin Delayed Deep Deterministic policy gradient (MATD3) with Hybrid Action Space and constructe a latent representation space for each agent to allow the agent to learn the latent policy. This enabling each AirX to act as an agent and autonomously optimize its hybrid action decisions to improve system efficiency in real-time based on the dynamic network environment, a capability not achievable by conventional DRL methods. This includes continuous optimization variables such as AirX deployment (location changes) and resource allocation, as well as discrete optimization variables for collaborative task offloading decisions. Extensive experiments against state-of-the-art algorithms (e.g., MADDPG, QMIX) demonstrate that the proposed system improves energy efficiency by 27.2% and task completion rate by 6.8% compared to traditional Integrated Ground-Air-Space (IG) Network.

The growing demand for intelligent mobile applications has made the deployment and operation of Deep Neural Networks (DNNs) on mobile Edge Devices (EDs) increasingly essential. However, the limited computational resources of EDs often result in significant energy consumption and compromised inference quality. To address these challenges, we propose a Partitioned Collaborative Inference (PCI) system that reduces on-device model inference costs by distributing the inference process across multiple EDs and MEC servers. To dynamically model the relationships between computing nodes, inference tasks, and resources, we employ Graph Neural Networks to construct the current state representation of the system. Furthermore, we develop a Cross-Entropy Method (CEM) based Evolutionary Reinforcement Learning algorithm, which leverages negative temporal difference (TD) error as a population fitness metric to generate elite individuals. The elite produces high-quality samples to improve learning efficiency, thereby obtaining optimal partitioned collaborative inference decisions and resource allocation in highly dynamic and complex search spaces. Extensive simulations demonstrate that the proposed approach significantly outperforms existing methods and benchmark schemes, achieving a 57. 5% increase in the inference task completion rate and a 65.7% reduction in system costs.

Federated learning may protect user privacy, reduce the transmission of a large amount of raw data, and is more compatible with smart home applications. Current federated learning faces two major problems including non-independent and identically (Non-IID) distributed data and high communication overhead. Personalized federated learning is a good method to deal with Non-IID data, but current personalized federated learning methods overlook the shared features of users' living habits in the same region. Hierarchical federated learning can reduce traffic on the core network, but its potential for personalization for smart home applications has not been considered. Therefore, to address these issues simultaneously, we propose hierarchical personalized federated learning. Specifically, we adopt a three-layer federated learning architecture of cloud-edge-client. On this basis, we use differential learning classification loss (DLCL), hierarchical balance loss (HBL) and balanced edge data selection (BEDS) methods to achieve the personalization of models on both the device side and the edge side. Finally, our experiments demonstrate that compared to state-of-the-art federated learning methods, hierarchical personalized federated learning has improvements in model accuracy and communication overhead.

Unmanned aerial vehicle (UAV) edge computing systems provide easy-to-deploy and low-cost services at those areas with inadequate infrastructure by deploying UAVs as moving edge servers for large-scale users. However, user devices are generally distributed unevenly in a large area, which makes it difficult for existing efforts to cope with this realistic scenario for optimal deployment of UAVs. Therefore, this paper considers a multiple UAV (Multi-UAV) Collaborative edge Computing (UCC) system by utilizing collaboration among them to split computation tasks at UAVs to balance the load and improve resource utilization. In order to maximize the energy-efficiency of the UCC system under the satisfaction of the delay constraint, we study the joint problem of UAV deployment, task collaborative offloading, computation and communication resource allocation in UCC system. We propose a bi-level optimization framework to solve the formulated non-convex mixed-integer optimization problem. In the upper level, the UAV deployment is optimized based on an improved differential evolution (DE) algorithm, and in the lower level the offloading decision and resource allocation are optimized based on a Reinforcement Learning (RL) algorithm with Twin Delayed Deep Deterministic policy gradient. Experimental results demonstrate the effectiveness and superiority of multi-UAV collaborative computing, with the proposed framework achieving a 32.4% reduction in energy consumption and an average 30% increase in task completion rate compared to DDPG, ToDeTaS, and other benchmark schemes.

The joint problem of task offloading, collaborative computing, and resource allocation for multi-access edge computing (MEC) is a challenging issue. In this article, splitting computing tasks at MEC servers through collaboration among MEC servers and a cloud server, we investigate the joint problem of collaborative task offloading and resource allocation. A collaborative task offloading, computing resource allocation, and subcarrier and power allocation problem in MEC is formulated. The goal is to minimize the total energy consumption of the MEC system while satisfying a delay constraint. The formulated problem is a nonconvex mixed-integer optimization problem. In order to solve the problem, we propose a deep reinforcement learning (DRL)-based bilevel optimization framework. The task offloading decision, computing collaboration decision, and power and subcarriers allocation subproblems are solved at the upper level, whereas the computing resource allocation subproblem is solved at the lower level. We combine dueling-DQN and double-DQN and add adaptive parameter space noise to improve DRL performance in MEC. Simulation results demonstrate that the proposed algorithm achieves near-optimal performance in energy efficiency and task completion rate compared with other DRL-based approaches and other benchmark schemes under various network parameter settings.

Mobile edge computing (MEC) is an emergent architecture, which brings computation and storage resources to the edge of mobile network and provides rich services and applications near the end users. The joint problem of task offloading and resource allocation in the multi-user collaborative mobile edge computing network (C-MEC) based on Orthogonal Frequency-Division Multiple Access (OFDMA) is a challenging issue. In this paper, we investigate the offloading decision, collaboration decision, computing resource allocation and communication resource allocation problem in C-MEC. The delay-sensitive tasks of users can be computed locally, offloaded to collaborative devices or MEC servers. The goal is to minimize the total energy consumption of all mobile users under the delay constraint. The problem is formulated as a mixed-integer nonlinear programming (MINLP), which involves the joint optimization of task offloading decision, collaboration decision, subcarrier and power allocation, and computing resource allocation. A two-level alternation method framework is proposed to solve the formulated MINLP problem. In the upper level, a heuristic algorithm is used to handle the collaboration decision and offloading decisions under the initial setting; and in the lower level, the allocation of power, subcarrier, and computing resources is updated through deep reinforcement learning based on the current offloading decision. Simulation results show that the proposed algorithm achieves excellent performance in energy efficient and task completion rate (CR) for different network parameter settings.