Large Language Models (LLMs) have demonstrated remarkable capabilities, leading to a significant increase in user demand for LLM services. However, cloud-based LLM services often suffer from high latency, unstable responsiveness, and privacy concerns. Therefore, multiple LLMs are usually deployed at the network edge to boost real-time responsiveness and protect data privacy, particularly for many emerging smart mobile and IoT applications. Given the varying response quality and latency of LLM services, a critical issue is how to route user requests from mobile and IoT devices to an appropriate LLM service (i.e., edge LLM expert) to ensure acceptable quality-of-service (QoS). Existing routing algorithms fail to simultaneously address the heterogeneity of LLM services, the interference among requests, and the dynamic workloads necessary for maintaining long-term stable QoS. To meet these challenges, in this paper we propose a novel deep reinforcement learning (DRL)-based QoS-aware LLM routing framework for sustained high-quality LLM services. Due to the dynamic nature of the global state, we propose a dynamic state abstraction technique to compactly represent global state features with a heterogeneous graph attention network (HAN). Additionally, we introduce an action impact estimator and a tailored reward function to guide the DRL agent in maximizing QoS and preventing latency violations. Extensive experiments on both Poisson and real-world workloads demonstrate that our proposed algorithm significantly improves average QoS and computing resource efficiency compared to existing baselines.

Graph-structured data applications promote the development of Graph Neural Networks (GNN) in recent years. Due to privacy concerns, collecting graph data stored in massive client devices for centralized graph learning is prohibitive. It is natural to integrate federated learning (FL) in graph learning to address this issue, which enables clients to collaborate on training a shared model without uploading their data. This generates an emerging paradigm of federated graph learning (FGL). However, due to various costs incurred by FGL training, the collaboration between the server and clients is still a challenging issue in FGL, which remains largely unexplored in existing studies. To bridge this gap, we propose a cost-efficient collaboration framework for FGL of graph convolutional networks on semi-supervised node classification tasks, i.e., Joint Client and Cross-Client Edge Selection (JC3ES) for the server. Specifically, we first characterize how varies graph structure affect the final convergence performance of the FGL model. We then reveal the fundamental supermodular property in client selection. Based on this, we further devise an approximately optimal algorithm for the server and theoretically derive the performance gap between the proposed algorithm and the optimal solution. Extensive numerical evaluations show that our proposed algorithm achieves outstanding performance in cost-efficient collaboration for FGL on popular graph datasets.

Mobile traffic prediction plays a crucial role in enabling efficient network management and service provisioning. Traditional prediction approaches treat different mobile application services (such as Uber, Facebook, Twitter, etc) as isolated entities, neglecting potential correlation among them. Moreover, such isolated prediction methods necessitate the uploading of historical traffic data from all regions to forecast city-wide traffic, resulting in consuming substantial bandwidth resources and risking prediction failure in the event of data loss in specific regions. To address these challenges, we propose a novel Cross-service Attention-based Spatial-Temporal Graph Convolutional Network (CsASTGCN) for precise and communication-efficient multi-service mobile traffic prediction. Our methodology allows each mobile service to transmit the traffic data of only a fraction of regions for city-wide traffic prediction of all mobile services, which reduces the resource consumption caused by data transmission. Specifically, the sparse traffic data are initially transmitted to the cloud server and the masked graph autoencoder is utilized to roughly reconstruct the traffic volume for regions with missing data. Subsequently, a cross-service attention-based predictor is designed to calculate the data correlation among different mobile services within the same region. Considering the constantly emerging mobile services, we incorporate a novel model-based adaptive transfer learning scheme to extract valuable knowledge from the existing models and expedite the training of a new model for a new service without training from scratch, thereby enhancing the scalability of our framework. Extensive experiments conducted on a large-scale real-world mobile traffic dataset demonstrate that our model greatly outperforms the existing schemes, enhancing both the communication-efficiency and robustness of large-scale multi-service traffic prediction.

Federated Learning (FL) requires frequent exchange of model parameters, which leads to long communication delay, especially when the network environments of clients vary greatly. Moreover, the parameter server needs to wait for the slowest client (i.e., straggler, which may have the largest model size, lowest computing capability or worst network condition) to upload parameters, which may significantly degrade the communication efficiency. Commonly-used client selection methods such as partial client selection would lead to the waste of computing resources and weaken the generalization of the global model. To tackle this problem, along a different line, in this paper, we advocate the approach of model parameter dropout instead of client selection, and accordingly propose a novel framework of Federated learning scheme with Differential parameter Dropout (FedDD). FedDD consists of two key modules: dropout rate allocation and uploaded parameter selection, which will optimize the model parameter uploading ratios tailored to different clients' heterogeneous conditions and also select the proper set of important model parameters for uploading subject to clients' dropout rate constraints. Specifically, the dropout rate allocation is formulated as a convex optimization problem, taking system heterogeneity, data heterogeneity, and model heterogeneity among clients into consideration. The uploaded parameter selection strategy prioritizes on eliciting important parameters for uploading to speedup convergence. Furthermore, we theoretically analyze the convergence of the proposed FedDD scheme. Extensive performance evaluations demonstrate that the proposed FedDD scheme can achieve outstanding performances in both communication efficiency and model convergence, and also possesses a strong generalization capability to data of rare classes.

Federated learning (FL) has emerged as a promising paradigm that enables clients to collaboratively train a shared global model without uploading their local data. To alleviate the heterogeneous data quality among clients, artificial intelligence-generated content (AIGC) can be leveraged as a novel data synthesis technique for FL model performance enhancement. Due to various costs incurred by AIGC-empowered FL (e.g., costs of local model computation and data synthesis), however, clients are usually reluctant to participate in FL without adequate economic incentives, which leads to an unexplored critical issue for enabling AIGC-empowered FL. To fill this gap, we first devise a data quality assessment method for data samples generated by AIGC and rigorously analyze the convergence performance of FL model trained using a blend of authentic and AI-generated data samples. We then propose a data quality-aware incentive mechanism to encourage clients’ participation. In light of information asymmetry incurred by clients’ private multi-dimensional attributes, we investigate clients’ behavior patterns and derive the server's optimal incentive strategies to minimize server's cost in terms of both model accuracy loss and incentive payments for both complete and incomplete information scenarios. Numerical results demonstrate that our proposed mechanism exhibits highest training accuracy and reduces up to 53.34% of the server's cost with real-world datasets, compared with existing benchmark mechanisms.

Low Earth orbit (LEO) satellites enable complex Earth observation tasks (e.g.,remote sensing and cooperative monitoring) by leveraging large-scale satellite-generated Earth imageries and state-of-the-art machine learning (ML) techniques. However, due to restricted downlink bandwidth and spotty connectivity, it is infeasible for the satellites to transmit all the imageries to ground stations for ML model training. To address this issue, we use federated learning (FL) to mitigate the significant overhead of raw data transmission only by enabling model parameter exchange. Traditional FL requires a central server for model parameter aggregation, which is impractical for distributed LEO satellite constellation due to the difficulty of identifying a suitable central satellite. To tackle such challenge, we take the unique topological characteristics of the LEO satellite constellation to design a decentralized FL framework that enables efficient model aggregation in LEO satellite networks without a central server. The framework can avoid the reliability and communication bandwidth problems of the central server in centralized FL. To mitigate the straggler effect and address the statistical heterogeneity, we then propose a novel offloading framework for decentralized FL in LEO satellite networks to aid the collaboration among multiple satellites for resource sharing. Based on it, we derive a satellite-centric threshold-based offloading strategy and a system-wide greedy-based iterative offloading decision making algorithm, in order to achieve delay and accuracy optimization under the computation and communication power constraints. Theoretical analysis demonstrates that the proposed framework contributes to the high training performance of the global model. Extensive experiments based on realistic datasets show that the proposed framework can reduce the system delay by up to 41% on average and improve the global model accuracy by up to 9.39% compared with benchmark policies.

As a privacy-preserving distributed learning paradigm, federated learning (FL) enables multiple client devices to train a shared model without uploading their local data. To further enhance the privacy protection performance of FL, differential privacy (DP) has been successfully incorporated into FL systems to defend against privacy attacks from adversaries. In FL with DP, how to stimulate efficient client collaboration is vital for the FL server due to the privacy-preserving nature of DP and the heterogeneity of various costs (e.g., computation cost) of the participating clients. However, this kind of collaboration remains largely unexplored in existing works. To fill in this gap, we propose a novel analytical framework based on Stackelberg game to model the collaboration behaviors among clients and the server with reward allocation as incentive in FL with DP. We first conduct rigorous convergence analysis of FL with DP and reveal how clients’ multidimensional attributes would affect the convergence performance of FL model. Accordingly, we solve the Stackelberg game and derive the collaboration strategies for both clients and the server. We further devise an approximately optimal algorithm for the server to efficiently conduct the joint optimization of the client set selection, the number of global iterations, and the reward payment for the clients. Numerical evaluations using real-world datasets validate our theoretical analysis and corroborate the superior performance of the proposed solution.

Federated learning (FL) is a promising paradigm that enables collaboratively learning a shared model across massive clients while keeping the training data locally. However, for many existing FL systems, clients need to frequently exchange model parameters of large data size with the remote cloud server directly via wide-area networks (WAN), leading to significant communication overhead and long transmission time. To mitigate the communication bottleneck, we resort to the hierarchical federated learning paradigm of HiFL, which reaps the benefits of mobile edge computing and combines synchronous client-edge model aggregation and asynchronous edge-cloud model aggregation together to greatly reduce the traffic volumes of WAN transmissions. Specifically, we first analyze the convergence bound of HiFL theoretically and identify the key controllable factors for model performance improvement. We then advocate an enhanced design of HiFlash by innovatively integrating deep reinforcement learning based adaptive staleness control and heterogeneity-aware client-edge association strategy to boost the system efficiency and mitigate the staleness effect without compromising model accuracy. Extensive experiments corroborate the superior performance of HiFlash in model accuracy, communication reduction, and system efficiency.

The space-air-ground integrated network (SAGIN), one of the key technologies for next-generation mobile communication systems, can facilitate data transmission for users all over the world, especially in some remote areas where vast amounts of informative data are collected by Internet of remote things (IoRT) devices to support various data-driven artificial intelligence (AI) services. However, training AI models centrally with the assistance of SAGIN faces the challenges of highly constrained network topology, inefficient data transmission, and privacy issues. To tackle these challenges, we first propose a novel topology-aware federated learning framework for the SAGIN, namely Olive Branch Learning (OBL). Specifically, the IoRT devices in the ground layer leverage their private data to perform model training locally, while the air nodes in the air layer and the ring-structured low earth orbit (LEO) satellite constellation in the space layer are in charge of model aggregation (synchronization) at different scales. To further enhance communication efficiency and inference performance of OBL, an efficient Communication and Non-IID-aware Air node-Satellite Assignment (CNASA) algorithm is designed by taking the data class distribution of the air nodes as well as their geographic locations into account. Furthermore, we extend our OBL framework and CNASA algorithm to adapt to more complex multi-orbit satellite networks. We analyze the convergence of our OBL framework and conclude that the CNASA algorithm contributes to the fast convergence of the global model. Extensive experiments based on realistic datasets corroborate the superior performance of our algorithm over the benchmark policies.

Federated Learning (FL) has been a promising paradigm in distributed machine learning that enables in-situ model training and global model aggregation. While it can well preserve private data for end users, to apply it efficiently on IoT devices yet suffer from their inherent variants: their available computing resources are typically constrained, heterogeneous, and changing dynamically. Existing works deploy FL on IoT devices by pruning a sparse model or adopting a tiny counterpart, which alleviates the workload but may have negative impacts on model accuracy. To address these issues, we propose Eco-FL, a novel Edge Collaborative pipeline based Federated Learning framework. On the client side, each IoT device collaborates with trusted available devices in proximity to perform pipeline training, enabling local training acceleration with efficient augmented resource orchestration. On the server side, Eco-FL adopts a novel grouping-based hierarchical architecture that combines synchronous intra-group aggregation and asynchronous inter-group aggregation, where a heterogeneity-aware dynamic grouping strategy that jointly considers response latency and data distribution is developed. To tackle the resource fluctuation during the runtime, Eco-FL further applies an adaptive scheduling policy to judiciously adjust workload allocation and client grouping at different levels. Extensive experimental results using both prototype and simulation show that, compared to state-of-the-art methods, Eco-FL can upgrade the training accuracy by up to 26.3%, reduce the local training time by up to 61.5%, and improve the local training throughput by up to 2.6 ×.

With the rapid development of mobile cellular technologies and the increasing popularity of mobile and Internet of Things (IoT) devices, timely mobile traffic forecasting with high accuracy becomes more and more critical for proactive network service provisioning and efficient network resource allocation in smart cities. Traditional traffic forecasting methods mostly rely on time series prediction techniques, which fail to capture the complicated dynamic nature and spatial relations of mobile traffic demand. In this paper, we propose a novel deep learning framework, graph attention spatial-temporal network (GASTN), for accurate citywide mobile traffic forecasting, which can capture not only local geographical dependency but also distant inter-region relationship when considering spatial factor. Specifically, GASTN considers spatial correlation through our constructed spatial relation graph and utilizes structural recurrent neural networks to model the global near-far spatial relationships as well as the temporal dependencies. In the framework of GASTN, two attention mechanisms are designed to integrate different effects in a holistic way. Besides, in order to further enhance the prediction performance, we propose a collaborative global-local learning strategy for the training of GASTN, which takes full advantage of the knowledge from both the global model and local models for individual regions and enhance the effectiveness of our model. Extensive experiments on a large-scale real-world mobile traffic dataset demonstrate that our GASTN model dramatically outperforms the state-of-the-art methods. And it reveals that a significant enhancement in the prediction performance of GASTN can be obtained by leveraging the collaborative global-local learning strategy.

In-home health monitoring has attracted great attention for the ageing population worldwide. With the abundant user health data accessed by Internet of Things (IoT) devices and recent development in machine learning, smart healthcare has seen many successful stories. However, existing approaches for in-home health monitoring do not pay sufficient attention to user data privacy and thus are far from being ready for large-scale practical deployment. In this paper, we propose FedHome, a novel cloud-edge based federated learning framework for in-home health monitoring, which learns a shared global model in the cloud from multiple homes at the network edges and achieves data privacy protection by keeping user data locally. To cope with the imbalanced and non-IID distribution inherent in user’s monitoring data, we design a generative convolutional autoencoder (GCAE), which aims to achieve accurate and personalized health monitoring by refining the model with a generated class-balanced dataset from user’s personal data. Besides, GCAE is lightweight to transfer between the cloud and edges, which is useful to reduce the communication cost of federated learning in FedHome. Extensive experiments based on realistic human activity recognition data traces corroborate that FedHome significantly outperforms existing widely-adopted methods.

Precise citywide mobile traffic prediction is of great significance for intelligent network planning and proactive service provisioning. Current traffic prediction approaches mainly focus on training a well-performed model for the cities with a large amount of mobile traffic data. However, for the cities with scarce data, the prediction performance will be greatly limited. To tackle this problem, in this paper we propose a novel cross-city deep transfer learning framework named CCTP for citywide mobile traffic prediction in cities with data scarcity. Specifically, we first present a novel spatial-temporal learning model and pre-train the model by abundant data of a source city to obtain prior knowledge of mobile traffic dynamics. We then devise an efficient generative adversarial network (GAN) based cross-domain adapter for distribution alignment between target data and source data. To deal with data scarcity issue in some clusters of target city, we further design an inter-cluster transfer learning strategy for performance enhancement. Extensive experiments conducted on real-world mobile traffic datasets demonstrate that our proposed CCTP framework can achieve superior performance in citywide mobile traffic prediction with data scarcity.

To meet the ever increasing mobile traffic demand in 5G era, base stations (BSs) have been densely deployed in radio access networks (RANs) to increase the network coverage and capacity. However, as the high density of BSs is designed to accommodate peak traffic, it would consume an unnecessarily large amount of energy if BSs are on during off-peak time. To save the energy consumption of cellular networks, an effective way is to deactivate some idle base stations that do not serve any traffic demand. In this paper, we develop a traffic-aware dynamic BS sleep control framework, named DeepBSC, which presents a novel data-driven learning approach to determine the BS active/sleep modes while meeting lower energy consumption and satisfactory Quality of Service (QoS) requirements. Specifically, the traffic demands are predicted by the proposed GS-STN model, which leverages the geographical and semantic spatial-temporal correlations of mobile traffic. With accurate mobile traffic forecasting, the BS sleep control problem is cast as a Markov Decision Process that is solved by Actor-Critic reinforcement learning methods. To reduce the variance of cost estimation in the dynamic environment, we propose a benchmark transformation method that provides robust performance indicator for policy update. To expedite the training process, we adopt a Deep Deterministic Policy Gradient (DDPG) approach, together with an explorer network, which can strengthen the exploration further. Extensive experiments with a real-world dataset corroborate that our proposed framework significantly outperforms the existing methods.

Online social networks (OSNs) are emerging as the most popular mainstream platform for content cascade diffusion. In order to provide satisfactory quality of experience (QoE) for users in OSNs, much research dedicates to proactive content placement by using the propagation pattern, user's personal profiles and social relationships in open social network scenarios (e.g., Twitter and Weibo). In this paper, we take a new direction of popularity-aware content placement in a closed social network (e.g., WeChat Moment) where user's privacy is highly enhanced. We propose a novel data-driven holistic deep learning framework, namely DeepCP, for joint diffusion-aware cascade prediction and autonomous content placement without utilizing users' personal and social information. We first devise a time-window LSTM model for content popularity prediction and cascade geo-distribution estimation. Accordingly, we further propose a novel autonomous content placement mechanism CP-GAN which adopts the generative adversarial network (GAN) for agile placement decision making to reduce the content access latency and enhance users' QoE. We conduct extensive experiments using cascade diffusion traces in WeChat Moment (WM). Evaluation results corroborate that the proposed DeepCP framework can predict the content popularity with a high accuracy, generate efficient placement decision in a real-time manner, and achieve significant content access latency reduction over existing schemes.

Federated Learning (FL) has been proposed as an appealing approach to handle data privacy issue of mobile devices compared to conventional machine learning at the remote cloud with raw user data uploading. By leveraging edge servers as intermediaries to perform partial model aggregation in proximity and relieve core network transmission overhead, it enables great potentials in low-latency and energy-efficient FL. Hence we introduce a novel Hierarchical Federated Edge Learning (HFEL) framework in which model aggregation is partially migrated to edge servers from the cloud. We further formulate a joint computation and communication resource allocation and edge association problem for device users under HFEL framework to achieve global cost minimization. To solve the problem, we propose an efficient resource scheduling algorithm in the HFEL framework. It can be decomposed into two subproblems: resource allocation given a scheduled set of devices for each edge server and edge association of device users across all the edge servers. With the optimal policy of the convex resource allocation subproblem for a set of devices under a single edge server, an efficient edge association strategy can be achieved through iterative global cost reduction adjustment process, which is shown to converge to a stable system point. Extensive performance evaluations demonstrate that our HFEL framework outperforms the proposed benchmarks in global cost saving and achieves better training performance compared to conventional federated learning.

Recently, location prediction has attracted considerable research effort because of the popularity of location-based services, such as mobile advertising and recommendations. With the unprecedented proliferation of mobile social networks, such as WeChat and Twitter, we are able to use location service to bridge the online and offline worlds, which is of great significance to many smart city applications. Different from existing studies, in this paper, we promote a user-centric location prediction approach by leveraging a user's local mobile social information without involving other users' location privacy. We propose a factor graph learning model that integrates not only user's social and network information but also the correlations between a user's locations into a unified framework. Furthermore, we use ReliefF algorithm to select user-specific significant features for location prediction and define the measure of location entropy to study the similarity between location, network status, and social behavior. To show the benefit of precise location prediction, we further apply it to personalized service migration in mobile edge computing (MEC) and accordingly propose prediction-based amortizing algorithm and lazy migration algorithm that can well balance the tradeoff between migration cost and non-migration latency in a cost-efficient manner. We conduct extensive experiments using a real-world data trace, which shows that our model performs much better in location prediction compared with several classic methods and the MEC service quality can be significantly enhanced by leveraging the location prediction.

Edge computing (EC) has quickly ascended to be the de-facto standard for hosting emerging low-latency applications, as exemplified by intelligent video surveillance, Internet of Vehicles, and augmented reality. For EC, service function chaining is envisioned as a promising approach to configure various services in an agile, flexible, and cost-efficient manner. When running on top of geographically dispersed edge clouds, fully unleashing the benefits of service function chaining is, however, by no means trivial. In this paper, we propose an online orchestration framework for cross-edge service function chaining, which aims to maximize the holistic cost efficiency, via jointly optimizing the resource provisioning and traffic routing on-the-fly. This long-term cost minimization problem is difficult since it is NP-hard and involves future uncertain information. To simultaneously address these dual challenges, we carefully combine an online optimization technique with an approximate optimization method in a joint optimization framework, through: 1) decomposing the long-term problem into a series of one-shot fractional problem with a regularization technique and 2) rounding the fractional solution to a near-optimal integral solution with a randomized dependent scheme that preserves the solution feasibility. The resulting online algorithm achieves an outstanding performance guarantee, as verified by both rigorous theoretical analysis and extensive trace-driven simulations.

Recently, location prediction has attracted considerable research effort because of the popularity of location- based services, such as mobile advertising and recommendations. With the unprecedented proliferation of mobile social networks, we are able to use location service to bridge the online and offline worlds. Different from existing studies, in this paper we promote a user-centric location prediction approach by leveraging a user's local mobile social information without involving other users' location privacy. We propose a factor graph learning model that integrates not only user's social and network information, but also the correlations between user's locations into a unified framework. Furthermore, we use ReliefF algorithm to select user-specific significant features for location prediction and define the measure of location entropy to study the similarity between location, network status and social behavior. We conduct extensive experiments using a real-world dataset, which shows that our model performs much better in location prediction compared with several classic methods.