The unprecedented advancements in quantum technology have opened new prospects for the widespread adoption of quantum applications, placing new demands on the information transmission capabilities of large-scale quantum networks. Long-distance and stable entanglements are deemed as the lifeline in quantum network communication. However, some weaknesses, e.g., quantum decoherence, scarce quantum memory, and uneven-quality entanglement, of the quantum entanglement hinder the development. In this paper, we proposeSophon, an online transmission framework for quantum networks, which utilizes high-dimensional entanglements to concurrently transmit multi-qubit data to satisfy the transmission requirements of the real-time request set. We first model the quantum network with multi-qubit entanglement represented by$W$quantum state and then formulate the Entanglement Routing and Qubit Provisioning (ERQP) problem as a global-local optimization process. To solve theERQPproblem, we distributedly regard each network node as an RL agent for resource provisioning and extend the step-updating of the Markov Decision Process by introducing a centralized controller for entanglement route selection to optimize local and global objectives, respectively. Extensive simulations demonstrate, on the self-made simulation platform,Sophonachieves a$21.89\%-66.52\%$decrease in the communication cost, and is more robust on different scales of the network topology and the request set than the baselines.

In recent years, volumetric videos have gradually prospered as an intriguing video paradigm, offering users a fully immersive viewing experience with six Degrees of Freedom (DoF). However, most current live volumetric video streaming methods struggle to facilitate the real-time performance requirements due to the nature of frequent user interactions and the complexity of network environments during video playback. Inspired by the correlation between the human visual effects and adjacent frame motion features, we proposeCetus, a context-aware cross-layer coordination system for live volumetric videos. First, we present an application-layer Neural Radiance Fields (NeRF)-based codec framework that leverages spatio-temporal semantic information for optimizing the compression quality of each video frame. Second, we exploit a flexible cross-layer coordination framework that seamlessly integrates frame drop strategy with partially reliable transmission, orchestrating transport protocols and application-informed rates to enhance the Quality of Experience (QoE) for multiple users. Furthermore, we develop a lightweight branching decision tree algorithm that adaptively makes fine-grained frame drop decisions. Experimental evaluations of our implemented system prototype demonstrate that Cetus significantly outperforms existing baseline approaches. Compared to the state-of-the-art baselines, Cetus effectively improves video frame rate by at least 24.7% and video quality by an average of 32.6%.

Bandit with infinitely many arms (i.e.,X-armed bandit) is an important variant of multi-armed stochastic bandits, which is useful to model different networking problems under both wired and wireless settings, e.g., online caching, dynamic channel/power allocation, rate adaption. However, the problem becomes challenging when the characteristic of the networking setting is affected by the side information (i.e., context) and distributed behavior. In this paper, we identify and study a novel problem, decentralized contextualX-armed bandit, whereNagents collaboratively solve the problem within time spanT. The problem is nontrivial because the infinite arms challenge and statistical information consensus issue make our setting go beyond a simple combination ofX-armed bandits and multi-agent bandits. We develop a decentralized arm selection algorithm, called MACXUCB, by elaborating the contextual covering tree technique with a novelgossip communication protocol, which allows each agent to communicate efficiently with her neighbors. We prove that MACXUCB achieves a sublinear regret upper bound Õ(D1/2dX+2dY+4NdX+dY+1.5/dX+dY+2TdX+dY+1/dX+dY+2) given aggregation periodD, and the covering dimensionsdXanddYof arm and context spaces, which asymptomatically matches the lower bound Ω((DN)1/dX+dY+2TdX+dY+1/dX+dY+2) up to a time-dependent factor. Moreover, MACXUCB enjoys a sublinear communication complexity Õ(NDT0.5/(1–p)) when tuning parameterp∈ (0, 1/2). Finally, we carry out experiments to verify the performance of our MACXUCB. The results show the effectiveness and efficiency of our MACXUCB.

The use of estimators instead of stochastic gradients for updates has been shown to improve algorithm convergence rates of, but their impact on generalization remains under-explored. In this paper, we investigate how estimators influence generalization. Our focus is on two widely studied problems: stochastic optimization (SO) and stochastic compositional optimization (SCO), both under convex and non-convex settings. For SO problems, we first analyze the generalization error of the STORM algorithm as a foundational step. We then extend our analysis to SCO problems by introducing an algorithmic framework that encompasses several popular algorithmic approaches. Through this framework, we conduct a generalization analysis, uncovering new insights into the impact of estimators on generalization. Subsequently, we provide a detailed analysis of three specific algorithms within this framework: SCGD, SCSC, and COVER, to explore the effects of different estimator strategies. Furthermore, in the context of SCO, we propose a novel definition of stability and a new decomposition of excess risk in the non-convex setting. Our analysis indicates two key findings: (1) In SCO problems, eliminating the estimator for the gradient of the inner function does not impact generalization performance while significantly reducing computational and storage overhead. (2) Faster convergence rates are consistently associated with better generalization performance.

Electroencephalography (EEG) signals have emerged as a novel biometric feature in identity authentication. However, in highly sensitive scenarios such as remote access control and sensitive operation confirmation, identity authentication alone is insufficient to ensure system security. This paper proposes EEGAuth, an EEG-based secure and lightweight authentication system with cryptographic key generation, addressing the demand for integrated systems that enhance both security and user convenience by combining identity authentication and key generation into a unified solution. The proposed system employs a genetic algorithm for optimal channel selection, integrates a discrete wavelet transform with an autoencoder-based feature extraction framework, and implements a CNN-based architecture for robust identity authentication. In addition, the system discretizes feature vectors to generate unique and repeatable seeds, which are used as inputs to a secure hash function to produce keys. The evaluation results show that our model achieves a classification accuracy of 99.38% with only 15 channels, significantly outperforming state-of-the-art methods and baseline models. The generated cryptographic keys demonstrate robust security properties, as evidenced by their successful passage through NIST statistical test suite for randomness verification, scale index analysis for aperiodicity assessment, and autocorrelation testing for bit-sequence independence, collectively confirming their resistance to cryptographic attacks and compliance with security standards.

Industrial Internet of Things (IIoT) is an emerging technology that digitizes industrial production and realizes Industry 4.0. However, it shows that IIoT is difficult to enable sophisticated downstream applications without eliciting all devices to achieve collaborative intelligence. Existing works on IIoT either require the consolidation of various IIoT devices’ data into a single centralized server which has potential privacy breach, or coordinate devices to learn a global model in privacy-preserving federated learning (FL) but assume data across devices has the sample feature space and neglect the heterogeneity of IIoT devices. In this article, we propose Meta-computing-driven vertical FL (VFL) algorithms to achieve collaborative intelligence in IIoT where heterogeneous devices have imperfect data with incomplete features. Specifically, we first provide the modeling of N devices’ VFL to collectively train the submodels and the common model. We present the computing graph to clearly indicate the gradient evaluation. To enable a fast convergence performance, we design a variance-reduced gradient estimator that can be seamlessly integrated into the basic VFL. Finally, we evaluate our proposed VFL by conducting experiments on the MNIST dataset regarding image recognition and the DAWM dataset for detecting anomalies in wafer manufacturing. The experimental results show that our VFL for IIoT is both effective and efficient.

Federated learning (FL)has achieved state-of-the-art performance in distributed machine learning with privacy preservation, which promotes AIoT. However, FL is restricted by the expensive communication cost due to exchanging a large number of model parameters and model updates (e.g., gradients) between the aggregator and participants in multiple rounds. This could be challenging in resource-constrained networks where devices are often resource-constrained in terms of computation and communication. Existing works mainly focus on improving communication efficiency from local training and model/gradient compression; nevertheless, studying communication efficiency for FL from the perspective of gradient estimation remains unexplored. In this paper, we bridge this gap by conducting a systematic study on gradient estimation for the communication-efficient FL. We propose a bandit-based gradient estimation-aware FL ($\mathtt{BGEFL }$) framework that can directly estimate participants’ gradients with limited bandit feedback (i.e., their local function values). We prove that$\mathtt{BGEFL }$enjoys an$\mathcal {O}(1)$communication complexity, that is a constant-size uplink communication in which each client uploads only one point’s feedback in the uplink. Moreover, our bandit-based gradient estimator is communication-efficient, unbiased, and stable. We prove theconvergenceperformance of$\mathtt{BGEFL }$for training strongly convex, general convex, and non-convex models. Finally, we evaluate our$\mathtt{BGEFL }$over several datasets and the experimental results demonstrate the effectiveness of$\mathtt{BGEFL }$.

This paper delves into the realm of stochastic optimization for compositional minimax optimization—a pivotal challenge across various machine learning domains, including deep AUC and reinforcement learning policy evaluation. Despite its significance, the problem of compositional minimax optimization is still under-explored. Adding to the complexity, current methods of compositional minimax optimization are plagued by sub-optimal complexities or heavy reliance on sizable batch sizes. To respond to these constraints, this paper introduces a novel method, called Nested STOchastic Recursive Momentum (NSTORM), which can achieve the optimal sample complexity and obtain the nearly accuracy solution, matching the existing minimax methods. We also demonstrate that NSTORM can achieve the same sample complexity under the Polyak-Lojasiewicz (PL)-condition—an insightful extension of its capabilities. Yet, NSTORM encounters an issue with its requirement for low learning rates, potentially constraining its real-world applicability in machine learning. To overcome this hurdle, we present ADAptive NSTORM (ADA-NSTORM) with adaptive learning rates. We demonstrate that ADA-NSTORM can achieve the same sample complexity but the experimental results show its more effectiveness. All the proposed complexities indicate that our proposed methods can match lower bounds to existing minimax optimizations, without requiring a large batch size in each iteration. Extensive experiments support the efficiency of our proposed methods.

Microservice (MS) structures a service application as a collection of independently deployable service modules, making it particularly suitable for delivering complex applications in distributed computing systems. This paper investigates MS architecture over Mobile Edge Computing (MEC) networks (hereafter referred to as EdgeMS) and studies an EdgeMS placement problem that aims to deploy MS modules over the MEC network in a manner that maximizes the reward of MS application providers. A novel algorithm called Dual-GNN Deep Deterministic Policy Gradient (DG-DDPG) is proposed to establish an intelligent EdgeMS placement policy for optimizing the location of MS modules and performing fractional computing resource allocation. DG-DDPG leverages the graph neural network (GNN) to comprehend the graph-structured information encapsulated in the MS application structure and MEC network. A dual-GNN core is constructed in DG-DDPG, one GNN for MS applications to distill knowledge from intricate connections between MS modules, and the other GNN for MEC networks to capture complicated interactions between edge sites when providing EdgeMS. DG-DDPG embeds the dual-GNN core in a DDPG-based reinforcement learning framework, which not only handles temporal dependencies between EdgeMS placement decisions for maximizing long-term reward but also supports continuous action space for enabling fractional resource allocation. In particular, the learning process of DG-DDPG is tailored to address hard constraints (i.e., computing capacity and MS application completeness) in the EdgeMS placement problem. We design constraint-based regularization terms and add them to the objective of DG-DDPG, which facilitates the identification of feasible placement decisions during learning. We carry out systematic experiments to evaluate the performance of DG-DDPG, and the results show that DG-DDPG outperforms state-of-the-art benchmarks in terms of reward, service delay and deployment cost.

Federated Learning (FL)has achieved state-of-the-art performance in training a global model in a decentralized and privacy-preserving manner. Many recent works have demonstrated that incentive mechanism is of paramount importance for the success of FL. Existing incentives to FL either neglect communication efficiency, or consider communication efficiency but design the incentive mechanisms using non-cooperative games under complete information assumption, or study incentive mechanism under incomplete information but only apply to the sequential interaction setting. We shed light on this problem from the cooperative perspective and propose an incentive mechanism for communication-efficient FL based on the Nash bargaining theory. Specially, we formulate our incentive mechanism as a one-to-manyconcurrent bargaininggame among the aggregator and clients, and systematically analyze the Nash bargaining solution (NBS, game equilibrium) to design the incentive mechanism. It should be noted that the existingsequential bargainingis not suitable for incentivizing FL due to high (exponential) time complexity, which deteriorates the straggler problem in FL. Our formulated bargaining game is challenging due to the NP-hardness. We propose a probabilistic greedy-based client selection algorithm and derive an analytical payment solution as an approximate NBS. We prove the convergence guarantee of our incentive mechanism for communication-efficient FL. Finally, we conduct experiments over real-world datasets to evaluate the performance of our incentive mechanism.

Off-screen handwriting recognitionenriches the handwriting interaction paradigm for mobile devices. However, the existing approaches are only applicable to the specific environment and equipment conditions. In this paper, we proposeAcouWrite, a general, scalable and real-time handwriting recognition system based on active acoustic sensing. In detail, AcouWrite relies onactive acoustic sensingusing only a pair of microphones and speakers on the smartphone to capture real-time handwriting input. Particularly, we extract theshort-time dCIR (st-dCIR)to monitor the changes in the acoustic transmission channel resulting from finger movement. Technically, we use aCNN-GRUclassifier to complete the recognition task in AcouWrite. Moreover, we use data augmentation and spelling error correction methods to improve AcouWrite's robustness. To improve the generalization of our AcouWrite for new characters, we incorporate the transfer learning module into our AcouWrite. In various real-world environments, experiments demonstrate that AcouWrite achieves a mean recognition accuracy of 97.62%, a word accuracy (WA) of 96.4% and a character error rate (CER) of 1.5% for 100 common words, and an average response time of 94 milliseconds.

Image segmentation is an important problem in many safety-critical applications such as medical imaging and autonomous driving. Recent studies show that modern image segmentation models are vulnerable to adversarial perturbations, while existing attack methods mainly follow the idea of attacking image classification models. We argue that image segmentation and classification have inherent differences, and design an attack framework specially for image segmentation models. Our goal is to thoroughly explore the vulnerabilities of modern segmentation models, i.e., aiming to misclassify as many pixels as possible under a perturbation budget in both white-box and black-box settings.Our attack framework is inspired by certified radius, which was originally used by defenders to defend against adversarial perturbations to classification models. We are the first, from the attacker perspective, to leverage the properties of certified radius and propose a certified radius guided attack framework against image segmentation models. Specifically, we first adapt randomized smoothing, the state-of-the-art certification method for classification models, to derive the pixel’s certified radius. A larger certified radius of a pixel means the pixel is theoretically more robust to adversarial perturbations. This observation inspires us to focus more on disrupting pixels with relatively smaller certified radii. Accordingly, we design a pixel-wise certified radius guided loss, when plugged into any existing white-box attack, yields our certified radius-guided white-box attack.Next, we propose the first black-box attack to image segmentation models via bandit. A key challenge is no gradient information is available. To address it, we design a novel gradient estimator, based on bandit feedback, which is query-efficient and provably unbiased and stable. We use this gradient estimator to design a projected bandit gradient descent (PBGD) attack. We further use pixels’ certified radii and design a certified radius-guided PBGD (CR-PBGD) attack. We prove our PBGD and CR-PBGD attacks can achieve asymptotically optimal attack performance with an optimal rate. We evaluate our certified-radius guided white-box and black-box attacks on multiple modern image segmentation models and datasets. Our results validate the effectiveness of our certified radius-guided attack framework.

Federated learning (FL) has achieved state-of-the-art performance in distributed learning tasks with privacy requirements. However, it has been discovered that FL is vulnerable to adversarial attacks. The typical gradient inversion attacks primarily focus on attempting to obtain the client’s private input in a white-box manner, where the adversary is assumed to be either the client or the server. However, if both the clients and the server are honest and fully trusted, is the FL secure? In this paper, we propose a novel method called External Gradient Inversion Attack (EGIA) in the grey-box settings. Specifically, we concentrate on the point that public-shared gradients in FL are always transmitted through the intermediary nodes, which has been widely ignored. On this basis, we demonstrate that an external adversary can reconstruct the private input using gradients even if both the clients and the server are honest and fully trusted. We also provide a comprehensive theoretical analysis of the black-box attack scenario in which the adversary has only the gradients. We perform extensive experiments on multiple real-world datasets to test the effectiveness of EGIA. The outcomes of our experiments validate that the EGIA method is highly effective.

Federated learning (FL) has reshaped the learning paradigm by overcoming privacy concerns and siloed data. In FL, an aggregator schedules a set of mobile users (MUs) to collectively train a global model with their local datasets and subsequently aggregates their model updates. However, the users have many uncertainties like unstable network connections and volatile availability, which leads to the straggler problem and deteriorates the efficiency of the FL system. Besides, the issue of non-IID datasets hinders the convergence performance of the global model. To hurdle the user uncertainties, we associate a deadline with the decision in each round and partially collect MUs' updates after the deadline, which can be achieved by considering surplus budget constraints. Moreover, we introduce fairness constraints for the non-IID issue. We propose a deadline-aware task replication for surplus client scheduling policy, called FEDDATE-CS. FEDDATE-CS is developed based on a novel contextual-combinatorial multi-armed bandit (CCMAB) learning framework with fairness guarantee. We extend the hypercube-based CCMAB framework by integrating the Lyapunov queuing technique and rigorously prove that FEDDATE-CS achieves a sublinear regret bound and provides an$[\mathcal{O}(1/V),\mathcal{O}(V)]$regret-fairness tradeoff for any fairness control factor$V>0$. We conduct extensive evaluations to verify the significant superiority of FEDDATE-CS over benchmarks.

Graph neural networks (GNNs) have achieved state-of-the-art performance in many graph-based tasks such as node classification and graph classification. However, many recent works have demonstrated that an attacker can mislead GNN models by slightly perturbing the graph structure. Existing attacks to GNNs are either under the less practical threat model where the attacker is assumed to access the GNN model parameters, or under the practical black-box threat model but consider perturbing node features that are shown to be not enough effective. In this paper, we aim to bridge this gap and consider black-box attacks to GNNs with structure perturbation as well as with theoretical guarantees. We propose to address this challenge through bandit techniques. Specifically, we formulate our attack as an online optimization with bandit feedback. This original problem is essentially NP-hard due to the fact that perturbing the graph structure is a binary optimization problem. We then propose an online attack based on bandit optimization which is proven to be sublinear to the query number T, i.e., O(✓NT3/4) where N is the number of nodes in the graph. Finally, we evaluate our proposed attack by conducting experiments over multiple datasets and GNN models. The experimental results on various citation graphs and image graphs show that our attack is both effective and efficient.

Knowing model parameters has been regarded as a vital factor for recovering sensitive information from the gradients in federated learning. But is it safe to use federated learning when the model parameters are unavailable for adversaries, i.e., external adversaries’ In this paper, we answer this question by proposing a novel gradient inversion attack. Speciffically, we observe a widely ignored fact in federated learning that the participants’ gradient data are usually transmitted via the intermediary node. Based on this fact, we show that an external adversary is able to recover the private input from the gradients, even if it does not have the model parameters. Through extensive experiments based on several real-world datasets, we demonstrate that our proposed new attack can recover the input with pixelwise accuracy and feasible efficiency.

X-armed bandits have achieved the state-of-the-art performance in optimizing unknown stochastic continuous functions, which can model many machine learning tasks, specially in big data-driven personalized recommendation. However, bandit algorithms are vulnerable to adversarial attacks. Existing works mainly focus on attacking multi-armed bandits in discrete setting; nevertheless, the attacks against X-armed bandits in continuous setting have not been well explored. In this paper, we aim to bridge this gap and investigate the robustness problem for the X-armed bandits. Specifically, we consider data poisoning attack and propose an attack algorithm named Confidence Poisoning Attack algorithm, which could hijack the clean tree-based X-armed bandits algorithm, i.e., high confidence tree (HCT) and make it choose the nodes including the arm targeted by the attacker very frequently with a sub-linear attack cost, i.e., O(Tα)(0 <α< 1), where T is the total number of rounds. We evaluate the efficiency of our proposed attack algorithm through theoretical analysis and experiments.

As a continuous variant of Multi-armed bandits (MAB), $\mathcal{X}$-armed bandits have enriched many applications of online machine learning like personalized recommendation system. However, the attack and defense to the $\mathcal{X}$-armed bandits remain largely unexplored, though the MAB has proved to be vulnerable. In this paper, we aim to bridge this gap and investigate the robustness analysis for the $\mathcal{X}$-armed bandits. Specifically, we consider action-manipulation attack, which is practical but harder than the existing reward-manipulation attack. We propose an attack algorithm based on a lower bound tree (LBT), which can continuously hijack the learner’s action by perturbing $\mathcal{X}$-armed bandits’ high confidence tree (HCT) construction. As a result, the nodes including the arm targeted by the attacker is selected frequently with a sublinear attack cost. To defend against the LBT attack, we propose a robust version of the HCT algorithm, called RoHCT. We theoretically analyze that the regret of RoHCT is related to the upper bound of the total cost Q and still sublinear to total number of rounds T. We carry out experiments to evaluate the effectiveness of LBT and RoHCT.

Mobile crowdsensing (MCS) enables a platform to recruit users to collectively perform sensing tasks from requesters. In order to maximize the completion qualities of tasks, an incentive mechanism should be well designed for the platform to incentivize high-quality users’ participation. The existing works largely adopt the Stackelberg game to model the strategic interactions in the incentive mechanism. However, there are practical issues that are less investigated in the context of the Stackelberg-based incentive mechanism. First, the platform has no knowledge about users’ sensing qualities beforehand due to their private information. Second, the platform needs users’ continuous participation in the long run, which results in fairness requirements. Third, it is also crucial to protect users’ privacy due to the potential privacy leakage concerns (e.g., sensing qualities) after completing tasks. In this article, we jointly address these issues and propose the three-stage Stackelberg-based incentive mechanism for the platform to recruit participants. In detail, we leverage combinatorial volatile multiarmed bandits (CVMABs) to elicit unknown users’ sensing qualities. We use the drift-plus-penalty (DPP) technique in Lyapunov optimization to handle the fairness requirements. We blur the quality feedback with tunable Laplacian noise such that the incentive mechanism protects locally differential privacy (LDP). Finally, we carry out experiments to evaluate our incentive mechanism. The numerical results show that our incentive mechanism achievessublinearregret performance to learn unknown quality with fairness and privacy guarantee.

Bike Sharing Systems (BSSs) have emerged as an economical and eco-friendly solution to alleviate the last-mile problem in intelligent transport systems. As an exclusive route selection problem in BSSs, Bike Trip Planning (BTP) has a clear goal: to select the available routes with minimum time cost for bike users, while satisfying their basic needs, e.g., the bounded longest walking distance. In this paper, we propose a real-time Lyapunov-based Bike Trip Planning (LBTP) policy that considers the users’ waiting at stations, which has not been sufficiently studied in the literature. Both the total time cost and the service rate of all users are the core criteria of a BSS, so we make use of the Lyapunov optimization theory to make a tradeoff between them. Our policy can achieve self-organizing bike redistribution without extra redistribution budget required. The evaluation results show the superiority of our policy on the both system utility and user service rate compared with the existing BTP policies, and reveal the extra travel time fairness degree among different types of users under our policy.

Mobile Crowd Sensing (MCS) enables the platform to offer data-based service by incentivizing mobile users to perform sensing task and collecting sensing data from them. Most of the existing works on MCS only consider designing incentive mechanisms for a single MCS platform. In this paper, we study the incentive mechanism in MCS with multiple platforms under two scenarios: competitive platform and cooperative platform. We correspondingly propose new competitive and cooperative mechanisms for each scenario. In the competitive platform scenario, platforms decide their prices on rewards to attract more participants, while the users choose which platform to work for. We model such a competitive platform scenario as a two-stage Stackelberg game. In the cooperative platform scenario, platforms cooperate to share sensing data with each other. We model it as many-to-many bargaining. Moreover, we first prove the NP-hardness of exact bargaining and then propose heuristic bargaining. Finally, numerical results show that (1) platforms in the competitive platform scenario can guarantee their payoff by optimally pricing on rewards and participants can select the best platform to contribute; (2) platforms in the cooperative platform scenario can further improve their payoff by bargaining with other platforms for cooperatively sharing collected sensing data.

In this article, we propose points-of-interest (POI)-tagging App-assisted incentive mechanism (PTASIM), an incentive mechanism that explores the cooperation with POI-tagging App for mobile edge crowdsensing (MEC). PTASIM requests App to tag some edges to be POI, which further guides App users to perform tasks at that location. We further model the interactions of users, platform, and App by a three-stage decision process. App first determines the POI-tagging price to maximize its payoff. Platform and users subsequently decide how to determine tasks reward and select edges to be tagged, and how to select the best task to perform, respectively. We analyze the optimal solution in those stages. Specifically, we prove that greedy algorithm could provide the optimal solution for platform's payoff maximization in polynomial time. The numerical results show that: 1) the cooperation with App brings long-term and sufficient participation; and 2) the optimal strategies reduce platform's tasks cost as well as improve App's revenues.