Much of federated learning (FL) focuses on settings where local dataset statistics remain the same between training and testing. However, this assumption often does not hold in practice due to distribution shifts, motivating the development of domain generalization (DG) approaches that leverage source domain data to train models capable of generalizing to unseen target domains. In this paper, we are motivated by two major gaps in existing work on FL and DG: (1) the lack of formal mathematical analysis of DG objectives; and (2) DG research in FL being limited to the star-topology architecture. We develop Decentralized Federated Domain Generalization with Style Sharing ($\textit{StyleDDG}$), a decentralized DG algorithm which allows devices in a peer-to-peer network to achieve DG based on sharing style information inferred from their datasets. Additionally, we provide the first systematic approach to analyzing style-based DG training in decentralized networks. We cast existing centralized DG algorithms within our framework, and employ their formalisms to model $\textit{StyleDDG}$. We then obtain analytical conditions under which convergence of $\textit{StyleDDG}$ can be guaranteed. Through experiments on popular DG datasets, we demonstrate that $\textit{StyleDDG}$ can obtain significant improvements in accuracy across target domains with minimal communication overhead compared to baseline decentralized gradient methods.

Future wireless networks must support emerging applications where environmental awareness is as critical as data transmission. Integrated Sensing and Communication (ISAC) enables this vision by allowing base stations (BSs) to allocate bandwidth and power to mobile users (MUs) for communications and cooperative sensing. However, this resource allocation is highly challenging due to:(i)dynamic resource demands from MUs and resource supply from BSs, and(ii)the selfishness of MUs and BSs. To address these challenges, existing solutions rely on either real-time (online) resource trading, which incurs high overhead and failures, or static long-term (offline) resource contracts, which lack flexibility. To overcome these limitations, we propose theFuture Resource Bank for ISAC, a hybrid trading framework that integrates offline and online resource allocation through a level-wise client model, where MUs and their coalitions negotiate with BSs. We introduce two mechanisms:(i)Offline Role-Friendly Win-Win Matching (offRFW2M), leveraging overbooking to establish risk-aware, stable contracts, and(ii)Online Effective Backup Win-Win Matching (onEBW2M), which dynamically reallocates unmet demand and surplus supply. We theoretically prove stability, individual rationality, and weak Pareto optimality of these mechanisms. Through comprehensive experiments, we show that our framework improves social welfare, latency, and energy efficiency compared to existing methods.

Ensuring safe, robust, and scalable motion planning for multiagent systems in dynamic and uncertain environments is a persistent challenge, driven by complex interagent interactions, stochastic disturbances, and model uncertainties. To overcome these challenges, particularly the computational complexity of coupled decision-making and the need for proactive safety guarantees, we propose a reachability-enhanced dynamic potential game (RE-DPG) framework, which integrates game-theoretic coordination into reachability analysis. This approach formulates multiagent coordination as a dynamic potential game, where the Nash equilibrium (NE) defines optimal control strategies across agents. To enable scalability and decentralized execution, we develop a neighborhood-dominated iterative best response scheme, built upon an iterated$\varepsilon$-BR process that guarantees finite-step convergence to an$\varepsilon$-NE. This allows agents to compute strategies based on local interactions while ensuring theoretical convergence guarantees. Furthermore, to ensure safety under uncertainty, we integrate a multiagent forward reachable set mechanism into the cost function, explicitly modeling uncertainty propagation and enforcing collision avoidance constraints. Through both simulations and real-world experiments in 2-D and 3-D environments, we validate the effectiveness of RE-DPG across diverse operational scenarios.

Generative AI (GenAI) has emerged as a transformative technology, enabling customized and personalized AI-generated content (AIGC) services. In this paper, we address challenges of edge-enabled AIGC service provisioning, which remain underexplored in the literature. These services require executing GenAI models with billions of parameters, posing significant obstacles to resource-limited wireless edge. We subsequently introduce the formulation of joint model caching and resource allocation for AIGC services to balance a trade-off between AIGC quality and latency metrics. We obtain mathematical relationships of these metrics with the computational resources required by GenAI models via experimentation. Afterward, we decompose the formulation into a model caching subproblem on a long-timescale and a resource allocation subproblem on a short-timescale. Since the variables to be solved are discrete and continuous, respectively, we leverage a double deep Q-network (DDQN) algorithm to solve the former subproblem and propose a diffusion-based deep deterministic policy gradient (D3PG) algorithm to solve the latter. The proposed D3PG algorithm makes an innovative use of diffusion models as the actor network to determine optimal resource allocation decisions. Consequently, we integrate these two learning methods within the overarching two-timescale deep reinforcement learning (T2DRL) algorithm, the performance of which is studied through comparative numerical simulations.

Federated Learning (FL) offers a pioneering distributed learning paradigm that enables devices/clients to build a shared global model that can be obtained through frequent model transmissions between clients and a central server, causing high latency, energy consumption, and congestion over backhaul links. To overcome these drawbacks, Hierarchical Federated Learning (HFL) has emerged, which organizes clients into multiple clusters and utilizes edge nodes (e.g., edge servers) for intermediate model aggregations between clients and the central server. Current research on HFL mainly focus on enhancing model accuracy, latency, and energy consumption in scenarios with a stable/fixed set of clients. However, addressing the dynamic availability of clients – a critical aspect of real-world scenarios – remains underexplored. This study delves into optimizing client selection and client-to-edge associations in HFL under intermittent client participation so as to minimize overall system costs (i.e., delay and energy), while achieving fast model convergence. We unveil that achieving this goal involves solving a complex NP-hard problem. To tackle this, we propose a stagewise methodology that splits the solution into two stages, referred to as Plan A and Plan B. Plan A focuses on identifying long-term clients with high chance of participation in subsequent model training rounds. Plan B serves as a backup, selecting alternative clients when long-term clients are unavailable during model training rounds. This stagewise methodology offers a fresh perspective on client selection that can enhance both HFL and conventional FL via enabling low-overhead decision-making processes. Through evaluations on diverse datasets, we show that our methodology outperforms existing benchmarks on crucial factors such as model accuracy and system costs.

Existing studies on federated learning (FL) are mostly focused on system orchestration forstatic snapshotsof the network and makingstatic control decisions(e.g., spectrum allocation). However, real-world wireless networks are susceptible totemporal variationsof wireless channel capacity and users’ datasets. In this paper, we study the impacts of the dynamics: 1) wireless channels and 2) users’ datasets on the FL execution. The former is captured by introducing a set of discrete time events while the latter is characterized by a novelordinary differential equationand the metric ofdynamic model drift, formulated via apartial differential inequality, drawing concrete analytical connections between the dynamics of users’ datasets and FL accuracy. We then proposedynamiccooperative FLwith dedicatedMAC schedulers (DCLM), exploiting the unique features of open radio access network (O-RAN) to execute FL.DCLMentails: 1) a hierarchical device-to-device (D2D)-assisted model training; 2) dynamic control decisions through dedicated O-RAN MAC schedulers; and 3) asymmetric user selection. We provide extensive theoretical analysis to study the convergence ofDCLMand then aim to optimize its degrees of freedom (e.g., user selection and spectrum allocation) through a non-convex optimization problem. We develop a systematic and generic approach to obtain the solution for this problem. We finally show the efficiency ofDCLMvia numerical simulations and provide a series of future directions.

Federated learning (FL) is vulnerable to backdoor attacks, where adversaries alter model behavior on target classification labels by embedding triggers into data samples. While these attacks have received considerable attention in horizontal FL, they are less understood for vertical FL (VFL), where devices hold different features of the samples, and only the server holds the labels. In this work, we propose a novel backdoor attack on VFL which (i) does not rely on gradient information from the server and (ii) considers potential collusion among multiple adversaries for sample selection and trigger embedding. Our label inference model augments variational autoencoders with metric learning, which adversaries can train locally. A consensus process over the adversary graph topology determines which datapoints to poison. We further propose methods for trigger splitting across the adversaries, with an intensity-based implantation scheme skewing the server towards the trigger. Our convergence analysis reveals the impact of backdoor perturbations on VFL indicated by a stationarity gap for the trained model, which we verify empirically as well. We conduct experiments comparing our attack with recent backdoor VFL approaches, finding that ours obtains significantly higher success rates for the same main task performance despite not using server information. Additionally, our results verify the impact of collusion on attack performance.

Federated learning (FL) is a popular technique for distributing machine learning (ML) across a set of edge devices. In this paper, we study fully decentralized FL, where in addition to devices conducting training locally, they carry out model aggregations via cooperative consensus formation over device-to-device (D2D) networks. We introduce asynchronous, event-triggered communications among the devices to handle settings where access to a central server is not feasible. To account for the inherent resource heterogeneity and statistical diversity challenges in FL, we define personalized communication triggering conditions at each device that weigh the change in local model parameters against the available local network resources. We theoretically recover the$\mathcal {O}{(\ln {k} / \sqrt {k})}$convergence rate to the globally optimal model of decentralized gradient descent (DGD) methods in the setup of our methodology. We provide our convergence guarantees for the last iterates of models, under relaxed graph connectivity and data heterogeneity assumptions compared with the existing literature. To do so, we demonstrate a$B$-connected information flow guarantee in the presence of sporadic communications over the time-varying D2D graph. Our subsequent numerical evaluations demonstrate that our methodology obtains substantial improvements in convergence speed and/or communication savings compared to existing decentralized FL baselines.

A few recent studies have shown the benefits of using centrally pre-trained models to initialize federated learning (FL). However, existing methods do not generalize well when faced with an arbitrary set of downstream FL tasks. Specifically, they often (i) achieve limited accuracy, especially with unseen downstream labels, and (ii) result in significant accuracy variance, failing to provide a balanced performance across clients. To address these challenges, we propose CoPreFL, a collaborative/distributed pre-training approach that robustly initializes for downstream FL tasks. CoPreFL leverages model-agnostic meta-learning (MAML) that tailors the global model to mimic heterogeneous and unseen FL scenarios, resulting in a pre-trained model that is rapidly adaptable to any FL task. Our MAML procedure integrates performance variance into the meta-objective function, balancing performance across clients rather than solely optimizing for accuracy. Extensive experiments show that CoPreFL significantly enhances average accuracy and reduces variance in arbitrary downstream FL tasks with unseen/seen labels, outperforming various pre-training baselines. Additionally, CoPreFL proves compatible with different well-known FL algorithms used in downstream tasks, boosting performance in each case.

Over the past several years, various federated learning (FL) methodologies have been developed to improve model accuracy, a primary performance metric in machine learning. However, to utilize FL in practical decision-making scenarios, beyond considering accuracy, the trained model must also have a reliable confidence in each of its predictions, an aspect that has been largely overlooked in existing FL research. Motivated by this gap, we propose Non-Uniform Calibration for Federated Learning (NUCFL), a generic framework that integrates FL with the concept of model calibration. The inherent data heterogeneity in FL environments makes model calibration particularly difficult, as it must ensure reliability across diverse data distributions and client conditions. Our NUCFL addresses this challenge by dynamically adjusting the model calibration objectives based on statistical relationships between each client's local model and the global model in FL. In particular, NUCFL assesses the similarity between local and global model relationships, and controls the penalty term for the calibration loss during client-side local training. By doing so, NUCFL effectively aligns calibration needs for the global model in heterogeneous FL settings while not sacrificing accuracy. Extensive experiments show that NUCFL offers flexibility and effectiveness across various FL algorithms, enhancing accuracy as well as model calibration.

Decentralized federated learning (DFL) captures FL settings where both (i) model updates and (ii) model aggregations are exclusively carried out by the clients without a central server. Existing DFL works have mostly focused on settings where clients conduct a fixed number of local updates between local model exchanges, overlooking heterogeneity and dynamics in communication and computation capabilities. In this work, we propose Decentralized Sporadic Federated Learning ($\texttt{DSpodFL}$), a DFL methodology built on a generalized notion of *sporadicity* in both local gradient and aggregation processes. $\texttt{DSpodFL}$ subsumes many existing decentralized optimization methods under a unified algorithmic framework by modeling the per-iteration (i) occurrence of gradient descent at each client and (ii) exchange of models between client pairs as arbitrary indicator random variables, thus capturing *heterogeneous and time-varying* computation/communication scenarios. We analytically characterize the convergence behavior of $\texttt{DSpodFL}$ for both convex and non-convex models and for both constant and diminishing learning rates, under mild assumptions on the communication graph connectivity, data heterogeneity across clients, and gradient noises. We show how our bounds recover existing results from decentralized gradient descent as special cases. Experiments demonstrate that $\texttt{DSpodFL}$ consistently achieves improved training speeds compared with baselines under various system settings.

In an era, where the exponential growth of image data driven by the Internet of Things (IoT) is outpacing traditional storage solutions, this work explores and advances the potential of implicit neural representation (INR) as a transformative approach to image compression. INR leverages the function approximation capabilities of neural networks to represent various types of data. While previous research has employed INR to achieve compression by training small networks to reconstruct large images, no work has explored past the fundamental barrier of using one network per image. This work proposes a novel advancement by breaking this barrier and representing multiple images with a single network. By modifying the loss function during training, the proposed approach allows a small number of weights to represent a large number of images, even those significantly different from each other. A thorough analytical study of the convergence of this new training method is also carried out, establishing upper bounds that not only confirm the method’s validity but also offer insights into optimal hyperparameter design. The proposed method is evaluated on the Kodak, ImageNet, and CIFAR-10 datasets. Experimental results demonstrate that all 24 images in the Kodak dataset can be represented by linear combinations of two sets of weights, achieving a peak signal-to-noise ratio (PSNR) of 26.5 dB with as low as 0.2 bits per pixel (BPP). The proposed method matches the rate-distortion performance of state-of-the-art image codecs, such as BPG, on the CIFAR-10 dataset. Additionally, the proposed method maintains the fundamental properties of INR, such as arbitrary resolution reconstruction of images.

Although user cooperation cannot improve the capacity of Gaussian two-way channels (GTWCs) with independent noises, it can improve communication reliability. In this work, we aim to enhance and balance the communication reliability in GTWCs by minimizing the sum of error probabilities via joint design of encoders and decoders at the users. We first formulate general encoding/decoding functions, where the user cooperation is captured by the coupling of user encoding processes. The coupling effect renders the encoder/decoder design non-trivial, requiring effective decoding to capture this effect, as well as efficient power management at the encoders within power constraints. To address these challenges, we propose two different twoway coding strategies: linear coding and learning-based coding. For linear coding, we propose optimal linear decoding and discuss new insights on encoding regarding user cooperation to balance reliability. We then propose an efficient algorithm for joint encoder/decoder design. For learning-based coding, we introduce a novel recurrent neural network (RNN)-based coding architecture, where we propose interactive RNNs and a power control layer for encoding, and we incorporate bi-directional RNNs with an attention mechanism for decoding. Through simulations, we show that our two-way coding methodologies outperform conventional channel coding schemes (that do not utilize user cooperation) significantly in sum-error performance. We also demonstrate that our linear coding excels at high signal-to-noise ratios (SNRs), while our RNN-based coding performs best at low SNRs. We further investigate our two-way coding strategies in terms of power distribution, two-way coding benefit, different coding rates, and block-length gain.

The rapid advancement of Artificial Intelligence (AI) has introduced Deep Neural Network (DNN)-based tasks to the ecosystem of vehicular networks. These tasks are often computation-intensive, requiring substantial computation resources, which are beyond the capability of a single vehicle. To address this challenge, Vehicular Edge Computing (VEC) has emerged as a solution, offering computing services for DNN-based tasks through resource pooling via Vehicle-to-Vehicle/Infrastructure (V2V/V2I) communications. In this paper, we formulate the problem of joint DNN partitioning, task offloading, and resource allocation in VEC as a dynamic long-term optimization. Our objective is to minimize the DNN-based task completion time while guaranteeing the system stability over time. To this end, we first leverage a Lyapunov optimization technique to decouple the original long-term optimization with stability constraints into a per-slot deterministic problem. Afterwards, we propose a Multi-Agent Diffusion-based Deep Reinforcement Learning (MAD2RL) algorithm, incorporating the innovative use of diffusion models to determine the optimal DNN partitioning and task offloading decisions. Furthermore, we integrate convex optimization techniques into MAD2RL as a subroutine to allocate computation resources, enhancing the learning efficiency. Through simulations under real-world movement traces of vehicles, we demonstrate the superior performance of our proposed algorithm compared to existing benchmark solutions.

This paper explores an interesting worker recruitment challenge where the mobile crowd sensing and computing (MCSC) platform hires workers to complete tasks with varying quality requirements and budget limitations, amidst uncertainties in worker participation and local workloads. We propose an innovative hybrid worker recruitment framework that combines offline and online trading modes. The offline mode enables the platform to overbook long-term workers by pre-signing contracts, thereby managing dynamic service supply. This is modeled as a 0-1 integer linear programming (ILP) problem with probabilistic constraints on service quality and budget. To address the uncertainties that may prevent long-term workers from consistently meeting service quality standards, we also introduce an online temporary worker recruitment scheme as a contingency plan. This scheme ensures seamless service provisioning and is likewise formulated as a 0-1 ILP problem. To tackle these problems with NP-hardness, we develop three algorithms, namely,i)exhaustive searching,ii)unique index-based stochastic searching with risk-aware filter constraint,iii)geometric programming-based successive convex algorithm. These algorithms are implemented in a stagewise manner to achieve optimal or near-optimal solutions. Extensive experiments demonstrate our effectiveness in terms of service quality, time efficiency, etc.

The use of machine learning (ML) for cancer staging through medical image analysis has gained substantial interest across medical disciplines. When accompanied by the innovative federated learning (FL) framework, ML techniques can further overcome privacy concerns related to patient data exposure. Given the frequent presence of diverse data modalities within patient records, leveraging FL in a multi-modal learning framework holds considerable promise for cancer staging. However, existing works on multi-modal FL often presume that all data-collecting institutions have access to all data modalities. This oversimplified approach neglects institutions that have access to only a portion of data modalities within the system. In this work, we introduce a novel FL architecture designed to accommodate not only the heterogeneity of data samples, but also the inherent heterogeneity/non-uniformity of data modalities across institutions. We shed light on the challenges associated with varying convergence speeds observed across different data modalities within our FL system. Subsequently, we propose a solution to tackle these challenges by devising a distributed gradient blending and proximity-aware client weighting strategy tailored for multi-modal FL. To show the superiority of our method, we conduct experiments using The Cancer Genome Atlas program (TCGA) datalake considering different cancer types and three modalities of data: mRNA sequences, histopathological image data, and clinical information. Our results further unveil the impact and severity of class-based vs type-based heterogeneity across institutions on the model performance, which widens the perspective to the notion of data heterogeneity in multi-modal FL literature.

Vehicular clouds (VCs) play a crucial role in the Internet-of-Vehicles (IoV) ecosystem by securing essential computing resources for a wide range of tasks. This paPertackles the intricacies of resource provisioning in dynamic VCs for computation-intensive tasks, represented by undirected graphs for parallel processing over multiple vehicles. We model the dynamics of VCs by considering multiple factors, including varying communication quality among vehicles, fluctuating computing capabilities of vehicles, uncertain contact duration among vehicles, and dynamic data exchange costs between vehicles. Our primary goal is to obtain feasible assignments between task components and nearby vehicles, calledtemplates, in a timely manner with minimized task completion time and data exchange overhead. To achieve this, wepropose ahybrid graphtaskscheduling (P-HTS) methodology that combines offline and online decision-making modes. For the offline mode, we introduce an approach called risk-aware pilot isomorphic subgraph searching (RA-PilotISS), which predicts feasible solutions for task scheduling in advance based on historical information. Then, for the online mode, we propose time-efficient instantaneous isomorphic subgraph searching (TE-InstaISS), serving as a backup approach for quickly identifying new optimal scheduling template when the one identified by RA-PilotISS becomes invalid due to changing conditions. Through comprehensive experiments, we demonstrate the superiority of our proposed hybrid mechanism compared to state-of-the-art methods in terms of various evaluative metrics, e.g., time efficiency such as the delay caused by seeking for possible templates and task completion time, as well as cost function, upon considering different VC scales and graph task topologies.

Hierarchical Federated Learning (HFL) extends conventional Federated Learning (FL) by introducing intermedi ate aggregation layers, enabling distributed learning in geograph ically dispersed environments, particularly relevant for smart IoT systems, such as remote monitoring and battlefield operations, where cellular connectivity is limited. In these scenarios, UAVs serve as mobile aggregators, dynamically connecting terrestrial IoT devices. This paper investigates an HFL architecture with energy-constrained, dynamically deployed UAVs prone to communication disruptions. We propose a novel approach to minimize global training costs by formulating a joint optimization problem that integrates learning configuration, bandwidth allocation, and device-to-UAV association, ensuring timely global aggregation before UAV disconnections and redeployments. The problem accounts for dynamic IoT devices and intermittent UAV con nectivity and is NP-hard. To tackle this, we decompose it into three subproblems: (i) optimizing learning configuration and bandwidth allocation via an augmented Lagrangian to reduce training costs; (ii) introducing a device fitness score based on data heterogeneity (via Kullback-Leibler divergence), device-to UAV proximity, and computational resources, using a TD3-based algorithm for adaptive device-to-UAV assignment; (iii) developing a low-complexity two-stage greedy strategy for UAV redeployment and global aggregator selection, ensuring efficient aggregation despite UAV disconnections. Experiments on diverse real-world datasets validate the approach, demonstrating cost reduction and robust performance under communication disruptions.

Modern vehicular networks encounter a multitude of computation-intensive tasks that have unique processing topologies represented by graph structures. The integration of edge computing and vehicular networks has provided a unique platform for handling these tasks at the network edge. However, the complex structure of these tasks makes their scheduling and execution challenging. This paper proposes a Vehicular Computing-assisted Edge Network (VCEN) architecture, where graph tasks are scheduled over a Vehicle-Edge Collaborative Cloud (VECC) for parallel execution. Our goal is to obtain feasible mappings between task components and computing nodes in the VECC while minimizing task execution latency and energy consumption. We show that achieving this goal requires solving an NP-hard optimization problem with complex constraints related to task structure and VECC topology. We then propose a fast and lightweight approach for graph task scheduling over VECC that comprises two key phases. In the former phase, we introduce a preprocessing algorithm that reduces the graph task's dimensionality by merging important components and cutting redundant edges. In the latter phase, we deploy a cost-reduction-preferred mapping algorithm to obtain feasible mappings between task components and VECC. Through simulations, we demonstrate our superior performance in different network settings.

While network coverage maps continue to expand, many devices located in remote areas remain unconnected to terrestrial communication infrastructures, preventing them from getting access to the associated data-driven services. In this paper, we propose a cooperative ground-to-satellite federated learning (FL) methodology to facilitate machine learning service management over remote regions. Our methodology orchestrates satellite constellations to provide the following key functions during FL: (i) processing data offloaded from ground devices, (ii) aggregating models within device clusters, and (iii) relaying models/data to other satellites via inter-satellite links (ISLs). Due to the limited coverage time of each satellite over a particular remote area, we facilitate satellite transmission of trained models and acquired data to neighboring satellites via ISL, so that the incoming satellite can continue FL for the region. We also develop a training latency minimizer which optimizes over the amount of data to be offloaded from ground devices to satellites. Through experiments on benchmark datasets, we show that our scheme can significantly speed up the convergence of FL compared with terrestrial-only and other satellite baseline approaches.

Devices located in remote regions often lack coverage from well-developed terrestrial communication infrastructure. This not only prevents them from experiencing high quality communication services but also hinders the delivery of machine learning services in remote regions. In this paper, we propose a new federated learning (FL) methodology tailored to space-air-ground integrated networks (SAGINs) to tackle this issue. Our approach strategically leverages the nodes within space and air layers as both 1) edge computing units and 2) model aggregators during the FL process, addressing the challenges that arise from the limited computation powers of ground devices and the absence of terrestrial base stations in the target region. The key idea behind our methodology is the adaptive data offloading and handover procedures that incorporate various network dynamics in SAGINs, including the mobility, heterogeneous computation powers, and inconsistent coverage times of incoming satellites. We analyze the latency of our scheme and develop an adaptive data offloading optimizer, and also characterize the theoretical convergence bound of our proposed algorithm. Experimental results confirm the advantage of our SAGIN-assisted FL methodology in terms of training time and test accuracy compared with various baselines.

While network coverage maps continue to expand, many devices located in remote areas remain unconnected to terrestrial communication infrastructures, preventing them from getting access to the associated data-driven services. In this paper, we propose a ground-to-satellite cooperative federated learning (FL) methodology to facilitate machine learning service management over remote regions. Our methodology orchestrates satellite constellations to provide the following key functions during FL: (i) processing data offloaded from ground devices, (ii) aggregating models within device clusters, and (iii) relaying models/data to other satellites via inter-satellite links (ISLs). Due to the limited coverage time of each satellite over a particular remote area, we facilitate satellite transmission of trained models and acquired data to neighboring satellites via ISL, so that the incoming satellite can continue conducting FL for the region. We theoretically analyze the convergence behavior of our algorithm, and develop a training latency minimizer which optimizes over satellite-specific network resources, including the amount of data to be offloaded from ground devices to satellites and satellites’ computation speeds. Through experiments on three datasets, we show that our methodology can significantly speed up the convergence of FL compared with terrestrial-only and other satellite baseline approaches.

Radio access networks (RANs) in monolithic architectures have limited adaptability to supporting different network scenarios. Recently, open-RAN (O-RAN) techniques have begun adding enormous flexibility to RAN implementations. O-RAN is a natural architectural fit for cell-free massive multiple-input multiple-output (CFmMIMO) systems, where many geographically-distributed access points (APs) are employed to achieve ubiquitous coverage and enhanced user performance. In this paper, we address the decentralized pilot assignment (PA) problem for scalable O-RAN-based CFmMIMO systems. We propose a low-complexity PA scheme using a multi-agent deep reinforcement learning (MA-DRL) framework in which multiple learning agents perform distributed learning over the O-RAN communication architecture to suppress pilot contamination. Our approach does not require prior channel knowledge but instead relies on real-time interactions made with the environment during the learning procedure. In addition, we design a codebook search (CS) scheme that exploits the decentralization of our O-RAN CFmMIMO architecture, where different codebook sets can be utilized to further improve PA performance without any significant additional complexities. Numerical evaluations verify that our proposed scheme provides substantial computational scalability advantages and improvements in channel estimation performance compared to the state-of-the-art.

Vehicular Clouds (VCs) are modern platforms for processing of computation-intensive tasks over vehicles. Such tasks are often represented as Directed Acyclic Graphs (DAGs) consisting of interdependent vertices/subtasks and directed edges. However, efficient scheduling of DAG tasks over VCs presents significant challenges, mainly due to the dynamic service provisioning of vehicles within VCs and non-Euclidean representation of DAG tasks’ topologies. In this paper, we propose a Graph neural network-Augmented Deep Reinforcement Learning scheme (GA-DRL) for the timely scheduling of DAG tasks over dynamic VCs. In doing so, we first model the VC-assisted DAG task scheduling as a Markov decision process. We then adopt a multi-head Graph ATtention network (GAT) to extract the features of DAG subtasks. Our developed GAT enables a two-way aggregation of the topological information in a DAG task by simultaneously considering predecessors and successors of each subtask. We further introduce non-uniform DAG neighborhood sampling through codifying the scheduling priority of different subtasks, which makes our developed GAT generalizable to completely unseen DAG task topologies. Finally, we augment GAT into a double deep Q-network learning module to conduct subtask-to-vehicle assignment according to the extracted features of subtasks, while considering the dynamics and heterogeneity of the vehicles in VCs. Through simulating various DAG tasks under real-world movement traces of vehicles, we demonstrate that GA-DRL outperforms existing benchmarks in terms of DAG task completion time.

Vehicular cloud (VC) is a promising technology for processing computation-intensive applications (CI-Apps) on smart vehicles. Implementing VCs over the network edge faces two key challenges: (C1) On-board computing resources of a single vehicle are often insufficient to process a CI-App; (C2) The dynamics of available resources, caused by vehicles’ mobility, hinder reliable CI-App processing. This work is among the first to jointly address (C1) and (C2), while considering two common CI-App graph representations, directed acyclic graph (DAG) and undirected graph (UG). To address (C1), we consider partitioning a CI-App with$m$dependent (sub-)tasks into$k\le m$groups, which are dispersed across vehicles. To address (C2), we introduce a generalized reliability metric called conditional mean time to failure (C-MTTF). Subsequently, we increase the C-MTTF of dependent sub-tasks processing via introducing a general framework of redundancy-based processing of dependent sub-tasks over semi-dynamic VCs (RP-VC). We demonstrate thatRP-VCcan be modeled as a non-trivial semi-Markov process (SMP). To analyze this SMP model and its reliability, we develop a novel mathematical framework, called event stochastic algebra ($\langle e\rangle $-algebra). Based on$\langle e\rangle $-algebra, we propose decomposition theorem (DT) to transform the presented SMP to a decomposed SMP (D-SMP). We subsequently calculate the C-MTTF of our methodology. We demonstrate that$\langle e\rangle $-algebra and DT are general mathematical tools that can be used to analyze other cloud-based networks. Simulation results reveal the exactness of our analytical results and the efficiency of our methodology in terms of acceptance and success rates of CI-App processing.

Federated learning (FedL) has emerged as a popular technique for distributing model training over a set of wireless devices, via iterative local updates (at devices) and global aggregations (at the server). In this paper, we develop parallel successive learning (PSL), which expands the FedL architecture along three dimensions: (i) Network, allowing decentralized cooperation among the devices via device-to-device (D2D) communications. (ii) Heterogeneity, interpreted at three levels: (ii-a) Learning: PSL considers heterogeneous number of stochastic gradient descent iterations with different mini-batch sizes at the devices; (ii-b) Data: PSL presumes a dynamic environment with data arrival and departure, where the distributions of local datasets evolve over time, captured via a new metric for model/concept drift. (ii-c) Device: PSL considers devices with different computation and communication capabilities. (iii) Proximity, where devices have different distances to each other and the access point. PSL considers the realistic scenario where global aggregations are conducted with idle times in-between them for resource efficiency improvements, and incorporates data dispersion and model dispersion with local model condensation into FedL. Our analysis sheds light on the notion of cold vs. warmed up models, and model inertia in distributed machine learning. We then propose network-aware dynamic model tracking to optimize the model learning vs. resource efficiency tradeoff, which we show is an NP-hard signomial programming problem. We finally solve this problem through proposing a general optimization solver. Our numerical results reveal new findings on the interdependencies between the idle times in-between the global aggregations, model/concept drift, and D2D cooperation configuration.

The conventional federated learning (FedL) architecture distributes machine learning (ML) across worker devices by having them train local models that are periodically aggregated by a server. FedL ignores two important characteristics of contemporary wireless networks, however: (i) the network may contain heterogeneous communication/computation resources, and (ii) there may be significant overlaps in devices’ local data distributions. In this work, we develop a novel optimization methodology that jointly accounts for these factors via intelligent device sampling complemented by device-to-device (D2D) offloading. Our optimization methodology aims to select the best combination of sampled nodes and data offloading configuration to maximize FedL training accuracy while minimizing data processing and D2D communication resource consumption subject to realistic constraints on the network topology and device capabilities. Theoretical analysis of the D2D offloading subproblem leads to new FedL convergence bounds and an efficient sequential convex optimizer. Using these results, we develop a sampling methodology based on graph convolutional networks (GCNs) which learns the relationship between network attributes, sampled nodes, and D2D data offloading to maximize FedL accuracy. Through evaluation on popular datasets and real-world network measurements from our edge testbed, we find that our methodology outperforms popular device sampling methodologies from literature in terms of ML model performance, data processing overhead, and energy consumption.

By opportunistically engaging mobile users (workers), mobile crowdsensing (MCS) networks have emerged as important approach to facilitate sharing of sensed/gathered data of heterogeneous mobile devices. To assign tasks among workers and ensure low overheads, we introduce a series of stable matching mechanisms, which are integrated into a novel hybrid service trading paradigm consisting offutures tradingandspot tradingmodes, to ensure seamless MCS service provisioning. In futures trading, we determine a set of long-term workers for each task through anoverbooking-enabledin-advancemany-to-manymatching (OIA3M) mechanism, while characterizing the associated risks under statistical analysis. In spot trading, we investigate the impact of fluctuations in long-term workers' resources on the violation of service quality requirements of tasks, and formalize a spot trading mode for tasks with violated service quality requirements under practical budget constraints, where the task-worker mapping is carried out viaonsitemany-to-manymatching (O3M) andonsitemany-to-onematching (OMOM). We theoretically show that our proposed matching mechanisms satisfy stability, individual rationality, fairness, and computational efficiency. Comprehensive evaluations confirm the satisfaction of these properties in practical network settings and demonstrate our commendable performance in terms of service quality, running time, and decision-making overheads, e.g., delay and energy consumption.

Semi-decentralized federated learning blends the conventional device-to-server (D2S) interaction structure of federated model training with localized device-to-device (D2D) communications. We study this architecture over practical edge networks with multiple D2D clusters modeled as time-varying and directed communication graphs. Our investigation results in an algorithm that controls the fundamental trade-off between (a) the rate of convergence of the model training process towards the global optimizer, and (b) the number of D2S transmissions required for global aggregation. Specifically, in our semi-decentralized methodology, D2D consensus updates are injected into the federated averaging framework based on column-stochastic weight matrices that encapsulate the connectivity within the clusters. To arrive at our algorithm, we show how the expected optimality gap in the current global model depends on the greatest two singular values of the weighted adjacency matrices (and hence on the densities) of the D2D clusters. We then derive tight bounds on these singular values in terms of the node degrees of the D2D clusters, and we use the resulting expressions to design a threshold on the number of clients required to participate in any given global aggregation round so as to ensure a desired convergence rate. Simulations performed on real-world datasets reveal that our connectivity-aware algorithm reduces the total communication cost required to reach a target accuracy significantly compared with baselines depending on the connectivity structure and the learning task.

We investigate training machine learning (ML) models across a set of geo-distributed, resource-constrained clusters of devices through unmanned aerial vehicles (UAV) swarms. The presence of time-varying data heterogeneity and computational resource inadequacy among device clusters motivate four key parts of our methodology: (i)stratified UAV swarmsof leader, worker, and coordinator UAVs, (ii)hierarchical nested personalized federated learning(HN-PFL), a distributed ML framework for personalized model training across the worker-leader-core network hierarchy, (iii)cooperative UAV resource poolingto address computational inadequacy of devices by conducting model training among the UAV swarms, and (iv)model/concept driftto model time-varying data distributions. In doing so, we consider bothmicro(i.e., UAV-level) andmacro(i.e., swarm-level) system design. At the micro-level, we propose network-awareHN-PFL, where we distributively orchestrate UAVs inside swarms to optimize energy consumption and ML model performance with performance guarantees. At the macro-level, we focus on swarm trajectory and learning duration design, which we formulate as a sequential decision making problem tackled via deep reinforcement learning. Our simulations demonstrate the improvements achieved by our methodology in terms of ML performance, network resource savings, and swarm trajectory efficiency.

We propose cooperative edge-assisted dynamic federated learning (CE-FL).CE-FLintroduces a distributed machine learning (ML) architecture, where data collection is carried out at the end devices, while the model training is conducted cooperatively at the end devices and the edge servers, enabled via data offloading from the end devices to the edge servers through base stations.CE-FLalso introduces floating aggregation point, where the local models generated at the devices and the servers are aggregated at an edge server, which varies from one model training round to another to cope with the network evolution in terms of data distribution and users’ mobility.CE-FLconsiders the heterogeneity of network elements in terms of communication/computation models and the proximity to one another.CE-FLfurther presumes a dynamic environment with online variation of data at the network devices which causes a drift at the ML model performance. We model the processes taken duringCE-FL, and conduct analytical convergence analysis of its ML model training. We then formulate network-awareCE-FLwhich aims to adaptively optimize all the network elements via tuning their contribution to the learning process, which turns out to be a non-convex mixed integer problem. Motivated by the large scale of the system, we propose a distributed optimization solver to break down the computation of the solution across the network elements. We finally demonstrate the effectiveness of our framework with the data collected from a real-world testbed.

This article investigates vehicular cloud (VC)-assisted task scheduling in an air-ground integrated vehicular network (AGVN), where tasks carried by unmanned aerial vehicles (UAVs) and resources of VCs are both modeled as graph structures. We consider a scenario in which resource-limited UAVs carry a set of computation-intensive graph tasks, which are offloaded to resource-abundant vehicles for processing. We formulate an optimization problem to jointly optimize the mapping between task components and vehicles, and transmission powers of UAVs, while addressing the trade-off between i) completion time of tasks, ii) energy consumption of UAVs, and iii) data exchange cost among vehicles. We show that this problem is a mixed-integer non-linear programming, and thus NP-hard. We subsequently reveal that satisfying constraints related to graph task structure requires addressing the non-trivial subgraph isomorphism problem over a dynamic vehicular topology. Accordingly, we propose a decoupling approach by segregating template searching from transmission power allocation, where atemplatedenotes a mapping between task components and vehicles. For template search, we introduce a low-complexity algorithm for isomorphic subgraphs extraction. For power allocation, we develop an algorithm using$p$-norm and convex optimization techniques. Extensive simulations demonstrate that our approach outperforms baseline methods in various network settings.

We study the problem of learning representations that are private yet informative i.e., provide information about intended "ally" targets while hiding sensitive "adversary" attributes. We propose Exclusion-Inclusion Generative Adversarial Network (EIGAN), a generalized private representation learning (PRL) architecture that accounts for multiple ally and adversary attributes unlike existing PRL solutions. While centrally-aggregated dataset is a prerequisite for most PRL techniques, data in real-world is often siloed across multiple distributed nodes unwilling to share the raw data because of privacy concerns. We address this practical constraint by developing D-EIGAN, the first distributed PRL method that learns representations at each node without transmitting the source data. We theoretically analyze the behavior of adversaries under the optimal EIGAN and D-EIGAN encoders and the impact of dependencies among ally and adversary tasks on the optimization objective. Our experiments on various datasets demonstrate the advantages of EIGAN in terms of performance, robustness, and scalability. In particular, EIGAN outperforms the previous state-of-the-art by a significant accuracy margin ($47%$ improvement), and D-EIGAN’s performance is consistently on par with EIGAN under different network settings.

Traditional learning-based approaches to student modeling generalize poorly to underrepresented student groups due to biases in data availability. In this paper, we propose a methodology for predicting student performance from their online learning activities that optimizes inference accuracy over different demographic groups such as race and gender. Building upon recent foundations in federated learning, in our approach, personalized models for individual student subgroups are derived from a global model aggregated across all student models via meta-gradient updates that account for subgroup heterogeneity. To learn better representations of student activity, we augment our approach with a self-supervised behavioral pretraining methodology that leverages multiple modalities of student behavior (e.g., visits to lecture videos and participation on forums), and include a neural network attention mechanism in the model aggregation stage. Through experiments on three real-world datasets from online courses, we demonstrate that our approach obtains substantial improvements over existing student modeling baselines in predicting student learning outcomes for all subgroups. Visual analysis of the resulting student embeddings confirm that our personalization methodology indeed identifies different activity patterns within different subgroups, consistent with its stronger inference ability compared with the baselines.

Federated learning (FL) has been recognized as one of the most promising solutions for distributed machine learning (ML). In most of the current literature, FL has been studied for supervised ML tasks, in which edge devices collect labeled data. Nevertheless, in many applications, it is impractical to assume existence of labeled data across devices. To this end, we develop a novel methodology, Cooperative Federated unsupervised Contrastive Learning (CF-CL), for FL across edge devices with unlabeled datasets. CF-CL employs local device cooperation where data are exchanged among devices through device-to-device (D2D) communications to avoid local model bias resulting from non-independent and identically distributed (non-i.i.d.) local datasets. CF-CL introduces a push-pull smart data sharing mechanism tailored to unsupervised FL settings, in which, each device pushes a subset of its local datapoints to its neighbors as reserved datapoints, and pulls a set of datapoints from its neighbors, sampled through a probabilistic importance sampling technique. We demonstrate that CF-CL leads to (i) alignment of unsupervised learned latent spaces across devices, (ii) faster global convergence, allowing for less frequent global model aggregations; and (iii) is effective in extreme non-i.i.d. datasettings across the devices.

Intelligent reflecting surfaces (IRS) consist of configurable meta-atoms, which can alter the wireless propagation environment through design of their reflection coefficients. We consider adaptive IRS control in the practical setting where (i) the IRS reflection coefficients are attained by adjusting tunable elements embedded in the meta-atoms, (ii) the IRS reflection coefficients are affected by the incident angles of the incoming signals, (iii) the IRS is deployed in multi-path, time-varying channels, and (iv) the feedback link from the base station (BS) to the IRS has a low data rate. Conventional optimization-based IRS control protocols, which rely on channel estimation and conveying the optimized variables to the IRS, are not practical in this setting due to the difficulty of channel estimation and the low data rate of the feedback channel. To address these challenges, we develop a novel adaptive codebook-based limited feedback protocol to control the IRS. We propose two solutions for adaptive IRS codebook design: (i) random adjacency (RA), which utilizes correlations across the channel realizations, and (ii) deep neural network policy-based IRS control (DPIC), which is based on a deep reinforcement learning. Numerical evaluations show that the data rate and average data rate over one coherence time are improved substantially by the proposed schemes.

In this paper, we study a new latency optimization problem for blockchain-based federated learning (BFL) in multi-server edge computing. In this system model, distributed mobile devices (MDs) communicate with a set of edge servers (ESs) to handle both machine learning (ML) model training and block mining simultaneously. To assist the ML model training for resource-constrained MDs, we develop an offloading strategy that enables MDs to transmit their data to one of the associated ESs. We then propose a new decentralized ML model aggregation solution at the edge layer based on a consensus mechanism to build a global ML model via peer-to-peer (P2P)-based blockchain communications. Blockchain builds trust among MDs and ESs to facilitate reliable ML model sharing and cooperative consensus formation, and enables rapid elimination of manipulated models caused by poisoning attacks. We formulate latency-aware BFL as an optimization aiming to minimize the system latency via joint consideration of the data offloading decisions, MDs’ transmit power, channel bandwidth allocation for MDs’ data offloading, MDs’ computational allocation, and hash power allocation. Given the mixed action space of discrete offloading and continuous allocation variables, we propose a novel deep reinforcement learning scheme with a parameterized advantage actor critic algorithm. We theoretically characterize the convergence properties of BFL in terms of the aggregation delay, mini-batch size, and number of P2P communication rounds. Our numerical evaluation demonstrates the superiority of our proposed scheme over baselines in terms of model training efficiency, convergence rate, system latency, and robustness against model poisoning attacks.

The combinatorial auction (CA) is an efficient mechanism for resource allocation in different fields, including cloud computing. It can obtain high economic efficiency and user flexibility by allowing bidders to submit bids for combinations of different items instead of only for individual items. However, the problem of allocating items among the bidders to maximize the auctioneers’ revenue, i.e., the winner determination problem (WDP), is NP-complete to solve and inapproximable. Existing works for WDPs are generally based on mathematical optimization techniques and most of them focus on the single-unit WDP, where each item only has one unit. On the contrary, few works consider the multi-unit WDP in which each item may have multiple units. Given that the multi-unit WDP is more complicated but prevalent in cloud computing, we propose leveraging machine learning (ML) techniques to develop a novel low-complexity algorithm for solving this problem with negligible revenue loss. Specifically, we model the multi-unit WDP as an augmented bipartite bid-item graph and use a graph neural network (GNN) with half-convolution operations to learn the probability of each bid belonging to the optimal allocation. To improve the sample generation efficiency and decrease the number of needed labeled instances, we propose two different sample generation processes. We also develop two novel graph-based post-processing algorithms to transform the outputs of the GNN into feasible solutions. Through simulations on both synthetic instances and a specific virtual machine (VM) allocation problem in a cloud computing platform, we validate that our proposed method can approach optimal performance with low complexity and has good generalization ability in terms of problem size and user-type distribution.

Vehicular cloud computing has been emerged as a promising solution to fulfill users’ demands on processing computation-intensive applications in modern driving environments. Such applications are commonly represented by graphs consisting of components and edges. However, encouraging vehicles to share resources poses significant challenges owing to users’ selfishness. In this paper, an auction-based graph job allocation problem is studied in vehicular cloud-assisted networks considering resource reutilization. Our goal is to map each buyer (component) to a feasible seller (virtual machine) while maximizing the buyers’ utility-of-service, which concerns the execution time and commission cost. First, we formulate the auction-based graph job allocation as a 0-1 integer programming (0-1 IP) problem. Then, a Vickrey-Clarke-Groves based payment rule is proposed which satisfies the desired economical properties, truthfulness and individual rationality. We face two challenges: 1) the abovementioned 0-1 IP problem is NP-hard; 2) one constraint associated with the IP problem poses addressing the subgraph isomorphism problem. Thus, obtaining the optimal solution is practically infeasible in large-scale networks. Motivated by which, we develop a structure-preserved matching algorithm by maximizing the utility-of-service-gain, and the corresponding payment rule which offers economical properties and low computation complexity. Extensive simulations demonstrate that the proposed algorithm outperforms the contrast methods considering various problem sizes.

Federated learning has generated significant interest, with nearly all works focused on a “star” topology where nodes/devices are each connected to a central server. We migrate away from this architecture and extend it through thenetworkdimension to the case where there are multiple layers of nodes between the end devices and the server. Specifically, we develop multi-stage hybrid federated learning (MH-FL), a hybrid of intra-and inter-layer model learning that considers the network as amulti-layer cluster-based structure.MH-FLconsiders thetopology structuresamong the nodes in the clusters, including local networks formed via device-to-device (D2D) communications, and presumes asemi-decentralized architecturefor federated learning. It orchestrates the devices at different network layers in a collaborative/cooperative manner (i.e., using D2D interactions) to formlocal consensuson the model parameters and combines it with multi-stage parameter relaying between layers of the tree-shaped hierarchy. We derive the upper bound of convergence forMH-FLwith respect to parameters of the network topology (e.g., the spectral radius) and the learning algorithm (e.g., the number of D2D rounds in different clusters). We obtain a set of policies for the D2D rounds at different clusters to guarantee either a finite optimality gap or convergence to the global optimum. We then develop a distributed control algorithm forMH-FLto tune the D2D rounds in each cluster over time to meet specific convergence criteria. Our experiments on real-world datasets verify our analytical results and demonstrate the advantages ofMH-FLin terms of resource utilization metrics.

Intelligent reflecting surfaces (IRS) consist of configurable meta-atoms, which can change the wireless propagation environment through design of their reflection coefficients. We consider a practical setting where (i) the IRS reflection coefficients are configured by adjusting tunable elements embedded in the meta-atoms, (ii) the IRS reflection coefficients are affected by the incident angles of the incoming signals, (iii) the IRS is deployed in multi-path, time-varying channels, and (iv) the feedback link from the base station to the IRS has a low data rate. Conventional optimization-based IRS control protocols, which rely on channel estimation and conveying the optimized variables to the IRS, are not applicable in this setting due to the difficulty of channel estimation and the low feedback rate. Therefore, we develop a novel adaptive codebook-based limited feedback protocol where only a codeword index is transferred to the IRS. We propose two solutions for adaptive codebook design, random adjacency (RA) and deep neural network policy-based IRS control (DPIC), both of which only require the end-to-end compound channels. We further develop several augmented schemes based on RA and DPIC. Numerical evaluations show that the data rate and average data rate over one coherence time are improved substantially by our schemes.

We investigate a transmission mechanism aiming to improve the data rate between a base station (BS) and a user equipment (UE) through deploying multiple relaying UAVs. We consider the effect of interference incurred by another established communication network, which makes our problem challenging and different from the state of the art. We aim to design the 3D trajectories and power allocation for the UAVs to maximize the data flow of the network while keeping the interference on the existing communication network below a threshold. We utilize the mobility feature of the UAVs to evade the (un)-intended interference caused by (un)-intentional interferers. To this end, we propose an alternating-maximization approach to jointly obtain the 3D trajectories and the UAVs transmission powers. We handle the 3D trajectory design by resorting to spectral graph theory and subsequently address the power allocation through convex optimization techniques. We also approach the problem from the intentional interferer’s perspective where smart jammers chase the UAVs to effectively degrade the data flow of the network. We also extend our work to the case for multiple UEs. Finally, we demonstrate the efficacy of our proposed method through extensive simulations.

Federated learning has emerged as a popular technique for distributing model training across the network edge. Its learning architecture is conventionally a star topology be-tween the devices and a central server. In this paper, we propose two timescale hybrid federated learning (TT-Hf),which migrates to a more distributed topology via device-to-device (D2D) communications. In TT-HF, local model training occurs at devices via successive gradient iterations, and the synchronization process occurs at two timescales: (i) macro-scale, where global aggregations are carried out via device-server interactions, and (ii) micro-scale, where local aggregations are carried out via D2D cooperative consensus formation in different device clusters. Our theoretical analysis reveals how device, cluster, and network-level parameters affect the convergence of TT-HF, and leads to a set of conditions under which a convergence rate of O(1/t) is guaranteed. Experimental results demonstrate the improvements in convergence and utilization that can be obtained by TT-HF over state-of-the-art federated learning baselines.

In general, reliable communication via multiple-input multiple-output (MIMO) orthogonal frequency division multiplexing (OFDM) requires accurate channel estimation at the receiver. The existing literature largely focuses on denoising methods for channel estimation that depend on either (i) channel analysis in the time-domain with prior channel knowledge or (ii) supervised learning techniques which require large prelabeled datasets for training. To address these limitations, we present a frequency-domain denoising method based on a reinforcement learning framework that does not need a priori channel knowledge and pre-labeled data. Our methodology includes a new successive channel denoising process based on channel curvature computation, for which we obtain a channel curvature magnitude threshold to identify unreliable channel estimates. Based on this process, we formulate the denoising mechanism as a Markov decision process, where we define the actions through a geometry-based channel estimation update, and the reward function based on a policy that reduces mean squared error (MSE). We then resort to Q-learning to update the channel estimates. Numerical results verify that our denoising algorithm can successfully mitigate noise in channel estimates. In particular, our algorithm provides a significant improvement over the practical least squares (LS) estimation method and provides performance that approaches that of the ideal linear minimum mean square error (LMMSE) estimation with perfect knowledge of channel statistics.

This paper considers unmanned aerial vehicle (UAV)-assisted infrastructure-to-vehicle (I2V) communication employing UAVs as relays to increase the throughput between a roadside unit (RSU) and a vehicular user equipment (VUE). We investigate the UAV position planning problem under both single UAV and multiple cooperative UAVs scenarios while considering the mobility of the VUE, aiming to maximize the data rate of the system. We first consider using a single UAV and prove that the single UAV position planning can be formulated as a convex optimization problem, and then obtain the optimal position of the UAV. Next, we investigate the multiple cooperative UAVs scenario and formulate the joint power control and position planning problem to improve the data rate of the system under a fixed total power consumption. Numerical simulations are provided to verify our theoretical results. Our findings highlight the effects of important system parameters, such as height, transmit power, and the number of UAVs, on the optimal UAV positioning and system performance.

The conventional federated learning (FedL) architecture distributes machine learning (ML) across worker devices by having them train local models that are periodically aggregated by a server. FedL ignores two important characteristics of contemporary wireless networks, however: (i) the network may contain heterogeneous communication/computation resources, while (ii) there may be significant overlaps in devices' local data distributions. In this work, we develop a novel optimization methodology that jointly accounts for these factors via intelligent device sampling complemented by device-to-device (D2D) offloading. Our optimization aims to select the best combination of sampled nodes and data offloading configuration to maximize FedL training accuracy subject to realistic constraints on the network topology and device capabilities. Theoretical analysis of the D2D offloading subproblem leads to new FedL convergence bounds and an efficient sequential convex optimizer. Using this result, we develop a sampling methodology based on graph convolutional networks (GCNs) which learns the relationship between network attributes, sampled nodes, and resulting offloading that maximizes FedL accuracy. Through evaluation on real-world datasets and network measurements from our IoT testbed, we find that our methodology while sampling less than 5% of all devices outperforms conventional FedL substantially both in terms of trained model accuracy and required resource utilization.

Federated learning has emerged as a popular technique for distributing machine learning (ML) model training across the wireless edge. In this paper, we proposetwo timescale hybrid federated learning(TT-HF), a semi-decentralized learning architecture that combines the conventional device-to-server communication paradigm for federated learning with device-to-device (D2D) communications for model training. InTT-HF, during each global aggregation interval, devices (i) perform multiple stochastic gradient descent iterations on their individual datasets, and (ii) aperiodically engage in consensus procedure of their model parameters through cooperative, distributed D2D communications within local clusters. With a new general definition of gradient diversity, we formally study the convergence behavior ofTT-HF, resulting in new convergence bounds for distributed ML. We leverage our convergence bounds to develop an adaptive control algorithm that tunes the step size, D2D communication rounds, and global aggregation period ofTT-HFover time to target a sublinear convergence rate of$\mathcal {O}(1/t)$while minimizing network resource utilization. Our subsequent experiments demonstrate thatTT-HFsignificantly outperforms the current art in federated learning in terms of model accuracy and/or network energy consumption in different scenarios where local device datasets exhibit statistical heterogeneity. Finally, our numerical evaluations demonstrate robustness against outages caused by fading channels, as well favorable performance with non-convex loss functions.

Federated learning has emerged recently as a promising solution for distributing machine learning tasks through modern networks of mobile devices. Recent studies have obtained lower bounds on the expected decrease in model loss that is achieved through each round of federated learning. However, convergence generally requires a large number of communication rounds, which induces delay in model training and is costly in terms of network resources. In this paper, we propose a fast-convergent federated learning algorithm, called$\mathsf {FOLB}$, which performs intelligent sampling of devices in each round of model training to optimize the expected convergence speed. We first theoretically characterize a lower bound on improvement that can be obtained in each round if devices are selected according to the expected improvement their local models will provide to the current global model. Then, we show that$\mathsf {FOLB}$obtains this bound through uniform sampling by weighting device updates according to their gradient information.$\mathsf {FOLB}$is able to handle both communication and computation heterogeneity of devices by adapting the aggregations according to estimates of device’s capabilities of contributing to the updates. We evaluate$\mathsf {FOLB}$in comparison with existing federated learning algorithms and experimentally show its improvement in trained model accuracy, convergence speed, and/or model stability across various machine learning tasks and datasets.

The contemporary literature on cloud resource allocation is mostly focused on studying the interactions between customers and cloud managers. Nevertheless, the recent growth in the customers’ demands and the emergence of private cloud providers (CPs) entice the cloud managers to rent extra resources from the CPs so as to handle their backlogged tasks and attract more customers. This also renders the interactions between the cloud managers and the CPs an important problem to study. In this paper, we investigate both interactions through a two-stage auction mechanism. For the interactions between customers and cloud managers, we adopt the options-based sequential auctions (OBSAs) to design the cloud resource allocation paradigm. As compared to existing works, our framework can handle customers with heterogeneous demands, provide truthfulness as the dominant strategy, enjoy a simple winner determination procedure, and preclude the delayed entrance issue. We also provide the performance analysis of the OBSAs, which is among the first in literature. Regarding the interactions between cloud managers and CPs, we propose two parallel markets for resource gathering, and capture the selfishness of the CPs by theiroffered prices. We conduct a comprehensive analysis of the two markets and identify the bidding strategies of the cloud managers.

With the ease of deployment, capabilities of evading the jammers and obscuring their existence, unmanned aerial vehicles (UAVs) are one of the most suitable candidates to perform surveillance. There exists a body of literature in which the inspectors follow a deterministic trajectory to conduct surveillance, which results in a predictable environment for malicious entities. Thus, introducing randomness to the surveillance is of particular interest. In this work, we propose a novel framework for stochastic UAV-assisted surveillance that i) inherently considers the battery constraints of the UAVs, ii) proposes random moving patterns modeled via random walks, and iii) adds another degree of randomness to the system via considering probabilistic inspections. We formulate the problem of interest, i.e., obtaining the energy-efficient random walk and inspection policies of the UAVs subject to probabilistic constraints on inspection criteria of the sites and battery consumption of the UAVs, which turns out to be signomial programming that is highly non-convex. To solve it, we propose a centralized and a distributed algorithm along with their performance guarantee. This work contributes to both UAV-assisted surveillance and classic random walk literature by designing random walks with random inspection policies on weighted graphs with energy limited random walkers.

The integration of unmanned aerial vehicles (UAVs) into the terrestrial communications networks with a variety of tasks is viewed as a key technology for 5G and beyond. In this work, we consider the uplink millimeter-wave (mmWave) transmission between a set of UAVs and a base station (BS), where the UAVs deploy uplink non-orthogonal multiple access (NOMA) in multiple clusters. Furthermore, the BS also serves its own desired ground user equipment (UE) in the presence of many other ground UEs associated with other cells, which share the same frequency band. Considering the limited energy budget of UAVs, we formulate an energy efficiency (EE) problem, and propose a solution aided by the Dinkelbach's algorithm and successive convex approximation (SCA). Using realistic air-to-ground (A2G) and terrestrial channel models, we assess the performance of the proposed algorithm under various circumstances (maximum transmit power for UAVs, quality-of-service (QoS) constraint for the desired UE, etc.), and identify the best use cases.

We consider the relaying application of unmanned aerial vehicles (UAVs), in which UAVs are placed between two transceivers (TRs) to increase the throughput of the system. Instead of studying the placement of UAVs as pursued in existing literature, we focus on investigating the placement of a jammer or a major source of interference on the ground to effectively degrade the performance of the system, which is measured by the maximum achievable data rate of transmission between the TRs. We demonstrate that the optimal placement of the jammer is in general a non-convex optimization problem, for which obtaining the solution directly is intractable. Afterward, using the inherent characteristics of the signal-to-interference ratio (SIR) expressions, we propose a tractable approach to find the optimal position of the jammer. Based on the proposed approach, we investigate the optimal positioning of the jammer in both dual-hop and multi-hop UAV relaying settings. Numerical simulations are provided to evaluate the performance of our proposed method.

Vehicular cloud computing has emerged as a promising paradigm for fulfilling user requirements in computation-intensive tasks in modern driving environments. In this paper, a novel framework of multi-task offloading over vehicular clouds (VCs) is introduced where tasks and VCs along with their internal connections are modeled as undirected weighted graphs. Aiming to achieve a trade-off between minimizing task completion time and data exchange costs, task components are efficiently mapped to available virtual machines in the related VCs. The problem is formulated as a non-linear integer programming problem, mainly under constraints of limited contact between vehicles as well as available resources, and addressed considering different problem sizes. In small size scenarios with a couple of tasks and service providers in a VC, we determine optimal solutions; in larger size cases, a connection-restricted random-matching-based subgraph isomorphism algorithm is proposed that presents low computational complexity. Evaluation of the proposed algorithms against greedy-based baseline methods is conducted via extensive simulations.

Graph jobs represent a wide variety of computation-intensive tasks in which computations are represented by graphs consisting of components (denoting either data sources or data processing) and edges (corresponding to data flows between the components). Recent years have witnessed dramatic growth in smart vehicles and computation-intensive graph jobs, which pose new challenges to the provision of efficient services related to the Internet of Vehicles. Fortunately, vehicular clouds (VCs) formed by a collection of vehicles, which allows jobs to be offloaded among vehicles, can substantially alleviate heavy onboard workloads and enable on-demand provisioning of computational resources. In this article, we present a novel framework for VCs that maps components of graph jobs to service providers via opportunistic vehicle-to-vehicle communication. Then, graph job allocation over VCs is formulated as a nonlinear integer programming with respect to vehicles' contact duration and available resources, aiming to minimize the job completion time and data exchange cost. The problem is addressed for two scenarios: 1) low-traffic and 2) rush-hour scenarios. For the former, we determine the optimal solutions for the problem. In the latter case, given the intractable computations for deriving feasible allocations, we propose a novel low complexity randomized graph job allocation mechanism by considering hierarchical tree-based subgraph isomorphism extraction. The evaluation of the performance of both optimal and proposed randomized algorithms with two greedy-based baseline methods is carried out through extensive simulations.

In the era of big-data, the jobs submitted to the clouds exhibit complicated structures represented by graphs, where the nodes denote the sub-tasks each of which can be accommodated at a slot in a server, while the edges indicate the communication constraints among the sub-tasks. We develop a framework for efficient allocation of graph jobs in geo-distributed cloud networks (GDCNs), explicitly considering the power consumption of the datacenters (DCs). We address the following two challenges arising in graph job allocation: i) the allocation problem belongs to NP-hard nonlinear integer programming; ii) the allocation requires solving the NP-complete sub-graph isomorphism problem, which is particularly cumbersome in large-scale GDCNs. We develop a suite of efficient solutions for GDCNs of various scales. For small-scale GDCNs, we propose an analytical approach based on convex programming. For medium-scale GDCNs, we develop a distributed allocation algorithm exploiting the processing power of DCs in parallel. Afterward, we provide a novel low-complexity (decentralized) sub-graph extraction method, based on which we introduce cloud crawlers aiming to extract allocations of good potentials for large-scale GDCNs. Given these suggested strategies, we further investigate strategy selection under both fixed and adaptive DC pricing schemes, and propose an online learning algorithm for each.

We consider unmanned aerial vehicle (UAV)-assisted wireless communication employing UAVs as relays to increase the throughput between a pair of transmitter and receiver. We focus on developing effective methods to position the UAV(s) in the presence of interference in the environment, the existence of which makes the problem non-trivial and our methodology different from the current art. We study the optimal position planning, which aims to maximize the (average) signal-to-interference-ratio (SIR) of the system, in the presence of: i) one major source of interference, ii) stochastic interference. For each scenario, we first consider utilizing a single UAV in the dual-hop relay mode and determine its optimal position. Afterward, multiple UAVs in the multi-hop relay mode are considered, for which we investigate two novel problems concerned with determining the optimal number of required UAVs and developing an optimal distributed position alignment method. Subsequently, we propose a cost-effective method that simultaneously minimizes the number of UAVs and determines their optimal positions to guarantee a certain (average) SIR of the system. Alternatively, for a given number of UAVs, we develop a fully distributed placement algorithm along with its performance guarantee. Numerical simulations are provided to evaluate the performance of our proposed methods.

The deployment of the unmanned aerial vehicle (UAV) has been foreseen as a promising technology for the next generation communication networks. The distance limitation imposed by the line of sight RF connection can be removed by using RF coverage from existing commercial cellular service. In this work, we consider a transmission mechanism that aims to improve the data rate between a terrestrial base station (BS) and user equipment (UE) through deploying multiple UAVs relaying the desired data flow. Considering the coexistence of this network with other established communication networks, we take into account the effect of interference, which is incurred by the existing nodes. Our primary goal is to optimize the three-dimensional (3D) trajectories and power allocation for the relaying UAVs to maximize the data flow while keeping the interference to existing nodes below a predefined threshold. An alternating-maximization strategy is proposed to solve the joint 3D trajectory design and power allocation for the relaying UAVs. To this end, we handle the information exchange within the network by resorting to spectral graph theory and subsequently address the power allocation through convex optimization techniques. Simulation results show that our approach can considerably improve the information flow while the interference threshold constraint is met.

We consider unmanned aerial vehicle (UAV)-assisted wireless communication employing UAVs as relay nodes to increase the throughput between a pair of transmitter and receiver. We focus on developing effective methods to position the UAV(s) in the sky in the presence of a major source of interference, the existence of which makes the problem non-trivial. First, we consider utilizing a single UAV, for which we develop a theoretical framework to determine its optimal position aiming to maximize the SIR of the system. To this end, we investigate the problem for three practical scenarios, in which the position of the UAV is: (i) vertically fixed, horizontally adjustable; (ii) horizontally fixed, vertically adjustable; (iii) both horizontally and vertically adjustable. Afterward, we consider employing multiple UAVs, for which we propose a cost-effective method that simultaneously minimizes the number of required UAVs and determines their optimal positions so as to guarantee a certain SIR of the system. We further develop a distributed placement algorithm, which can increase the SIR of the system given an arbitrary number of UAVs. Numerical simulations are provided to evaluate the performance of our proposed methods.

Vehicular ad-hoc networks (VANETs) have recently attracted a lot of attention due to their immense potentials and applications. Wide range of coverage and accessibility to end users make VANETs a good target for commercial companies. In this paper, we consider a scenario in which advertising companies aim to disseminate their advertisements in different areas of a city by utilizing VANETs infrastructure. These companies compete for renting the VANETs infrastructure to spread their advertisements. We partition the city map into different blocks, and consider a manager for all the blocks who is in charge of splitting the time between interested advertising companies. Each advertising company (AdC) is charged proportional to the allocated time. In order to find the best time splitting between AdCs, we propose a Stackelberg game scheme in which the block manager assigns the companies to the blocks and imposes the renting prices to different companies in order to maximize its own profit. Based on this, AdCs request the amount of time they desire to rent the infrastructure in order to maximize their utilities. To obtain the Stackelberg equilibrium of the game, a mixed integer nonlinear optimization problem is solved using the proposed optimal and sub-optimal algorithms. The simulation results demonstrate that the sub-optimal algorithm approaches the optimal one in performance with lower complexity.

The goal of opinion maximization is to maximize the positive view towards a product, an ideology or any entity among the individuals in social networks. So far, opinion maximization is mainly studied as finding a set of influential nodes for fast content dissemination in a social network. In this paper, we propose a novel approach to solve the problem, where opinion maximization is achieved through efficient information spreading. In our model, multiple sources inject information continuously into the network, while the regular nodes with heterogeneous social learning abilities spread the information to their acquaintances through gossip mechanism. One of the sources employs smart information spreading and the rest spread information randomly. We model the social interactions and evolution of opinions as a dynamic Bayesian network (DBN), using which the opinion maximization is formulated as a sequential decision problem. Since the problem is intractable, we develop multiple variants of centralized and decentralized algorithms to obtain approximate solutions. Through simulations in synthetic and real-world networks, we demonstrate two key results: 1) the proposed methods perform better than random spreading by a large margin, and 2) even though the smart source (that spreads the desired content) is unfavorably located in the network, it can outperform the contending random sources located at favorable positions.

Vehicular ad-hoc networks (VANETs) has recently attracted a lot of attention due to their great potentials for different applications such as collision avoidance, route finding and autonomous driving. A wide range of coverage and accessibility to the end users in VANETs make them a good target for commercial advertising. This paper addresses the problem of mobile advertising in VANETs. We consider a case where different advertisers compete for the VANET infrastructure. It is assumed that a city is partitioned into a grid of blocks and the central data center manager (CDM) sets the rental price for each block considering the geographical position and the predicted density of vehicles inside the block. The regret-based minimization method is adopted to tackle the problem in response to its dynamic nature. Regret bound of the proposed algorithm and its convergence to the best strategy are shown rigorously. Furthermore, a good potential of the proposed algorithm is revealed through simulations.

Determining the causality of abnormalities in a network is the prerequisite for developing countermeasures. In this paper, we focus on infection detection in heterogeneous networks. Given a snapshot of the network which demonstrates the condition of the nodes, the goal is to distinguish between random failures and epidemic scenarios. We model the network situation as a graph signal based on the nodes' status. Detection metrics motivated by graph signal processing are introduced for the infection detection problem in hand, and an effective algorithm is proposed to solve it. Simulation results indicate a dramatic improvement in terms of detection probability compared to the current state-of-the-art.

Vehicular ad-hoc networks (VANETs) have been an active area in the research community during the last decade with focus primarily on routing protocols, security aspects and safety. Recent advances in wireless communication and the inherent dynamic nature of VANETs provide excellent opportunity for advertisement dissemination. In this paper, we address the problem of dynamic advertising in VANETs. We consider a city divided into a grid, where the blocks have different vehicular densities that vary over time. Several advertising companies compete for the blocks to broadcast their advertisements in the network. The content dissemination in the network is controlled by a data management unit that receives requests from advertising companies for each block. To solve the problem of block allocation, we adapt the repeated auction scheme for the dynamic setting. Two new metrics are defined to better represent the real- world scenario and fairness in repeated auctions. We propose an algorithm which is a combination of adaptive linear prediction and nonparametric Bayesian belief update, enabling smart bidding and improving the utilities of the competing advertising companies significantly in the long- run. Through simulations, we show that the proposed algorithm achieves better performance than two baselines approaches.

With growing demands for cloud computing services, the idea of managing limited cloud resources for making a profit has arisen as an important problem. Auction theory is recently considered as a viable way to solve the problem of cloud resource allocation. In this paper, we consider a model for Cloud of Clouds Networks (CCNs) with different types of servers along with customers with heterogeneous demands, in which customers and cloud servers may join and leave the CCN at will. We propose an options-based sequential auction that not only provides a good match with the dynamic structure of the problem, but also solves the entrance time problem and possesses the truthfulness property. We study both first-price and second-price options-based sequential auctions, and model the price matching processes in those auctions as Markov chains. We provide mathematically tractable methods to find the expected value of the CCN manager's revenue, and further show how the proxy agents' patience time affects the CCN manager's revenue.

In this paper, we consider the optimal design of interlinks for an interdependent system of networks. In contrast to existing literature, we explicitly exploit the information of intra-layer node degrees to design interdependent structures such that their robustness against cascading failures, triggered by randomized attacks, is maximized. Utilizing percolation theory-based system equations relating the robustness of the network to its degree sequence, we characterize the optimal design for the one-to-one structure, with complete interdependence and partial interdependence, under randomized attack. We also extend our study to the one-to-many interdependence structure and the targeted attack model. The theoretically derived optimal interdependence structures have been verified using simulations on scale-free networks.