Estimating correspondences between pairs of non-rigid deformable 3D shapes remains a significant challenge in computer vision and graphics. While deep functional map methods have become the go-to solution for addressing this problem, they primarily focus on optimizing pointwise and functional maps either individually or jointly, rather than directly enhancing feature representations in the embedding space, which often results in inadequate feature quality and suboptimal matching performance. Furthermore, these approaches heavily rely on traditional functional map techniques, such as time-consuming functional map solvers, which incur substantial computational costs. In this work, we introduce, for the first time, a novel unsupervised contrastive learning-based approach for efficient and robust 3D shape matching. We begin by presenting an unsupervised contrastive learning framework that promotes feature learning by maximizing consistency within positive similarity pairs and minimizing it within negative similarity pairs, thereby improving both the consistency and discriminability of the learned features. We then design a significantly simplified functional map learning architecture that eliminates the need for computationally expensive functional map solvers and multiple auxiliary functional map losses, greatly enhancing computational efficiency. By integrating these two components into a unified two-branch pipeline, our method achieves state-of-the-art performance in both accuracy and efficiency. Extensive experiments demonstrate that our approach is not only computationally efficient but also outperforms current state-of-the-art methods across various challenging benchmarks, including near-isometric, non-isometric, and topologically inconsistent scenarios—even surpassing supervised techniques.

Logical reasoning is a key capability of large language models, yet current benchmarks focus almost entirely on tasks that just check basic logical consistency and overlook the reflective reasoning required for paradox detection and resolution.To fill the gap, we present ParaSuite, the first pipeline dedicated to paradox research that automates data synthesis, evaluation, and training.We introduce PARADOX, a synthetic, high-quality data spanning two difficulty tiers and three academic domains, accompanied by specialized evaluation metrics and solving algorithms.We propose ParadoxBreaker-7B, trained with Mutual-Information Guided Fine-Tuning and reinforcement learning step verify paradox reward(PAPO).Experiments demonstrate significant improvements in both paradoxical and general STEM reasoning.

Positional encodings are fundamental to Transformers, yet explicit methods like RoPE can degrade under length extrapolation and may incur extra arithmetic and memory-access overhead.In this paper, we propose Scoped Position Encoding (SCOPE), a novel framework that reimagines structured sparsity as an intrinsic position encoding mechanism.Instead of relying on explicit arithmetic signals, SCOPE assigns exponentially scaled look-back scopes to attention heads.We theoretically demonstrate that this simple topological constraint transforms multi-head attention into a hierarchical processor, yielding an order awareness horizon that grows exponentially with depth up to the sequence length.Consequently, SCOPE is parameter-free and avoids relying on fragile positional arithmetic.Empirically, it significantly enhances efficiency by masking the majority of attention computations, offering a theoretical 8× reduction in attention FLOPs at long contexts.Extensive evaluations on LLaMA-3-8B architectures reveal that SCOPE achieves superior native length extrapolation and robust retrieval fidelity compared to RoPE, all while substantially reducing training and inference latency.The code is available at https://github.com/oncemoe/ScoPE.

With the widespread adoption of large vision-language models, the capacity for color vision in these models is crucial. However, the color vision abilities of large visual-language models have not yet been thoroughly explored. To address this gap, we define a color vision testing task for large vision-language models and construct a dataset that covers multiple categories of test questions and tasks of varying difficulty levels. Furthermore, we analyze the types of errors made by large vision-language models and propose a chain-of-thought prompting strategy to enhance their performance in color vision tests.

The integrated satellite-terrestrial network architecture has emerged as a hotspot for the Internet of Everything, which is a promising approach to provide communication service anytime and anywhere. In this paper, we aim to facilitate the system delay reduction through the two-tier cooperative edge computing of terrestrial base stations (BSs) and the satellite. To avoid the two-way propagation delay brought by the centralized decision process, we propose a distributed offloading scheme for the BSs and the satellite. Firstly, we equivalently decompose the offloading problem into multiple subproblems, based on which the optimal User-BS time slot allocation strategy for each BS is derived. Then, by means of multi-agent deep deterministic policy gradient reinforcement learning, the resource allocation strategy of each BS is obtained distributedly, which effectively avoids the two-way propagation delay. Based on the BS resource allocation strategy obtained, the optimal satellite computation capacity allocation strategy is proposed to minimize the total edge computing delay at the satellite. Finally, exhaustive simulations are implemented to demonstrate the performance of the proposed distributed offloading strategy.

Renewable energy has become a viable alternative to fossil fuels owing to its environmental benefits. However, its inherent uncertainty pose significant challenges. Demand response mechanisms have been developed to address these issues, facilitating renewable energy integration through consumer-side flexible resources. However, these mechanisms often affect consumer satisfaction, necessitating precise measurement and control of these impacts. In this paper, we propose a two-stage electricity trading and load dispatch optimization model aimed at reducing carbon emission by promoting renewable energy accommodation, and the proposed optimization model takes into account multi-category energy consumer satisfaction. We begin by classifying consumers into distinct categories and designing tailored satisfaction functions that reflect their unique power consumption preferences. The electricity trading and load dispatch processes are formulated as a two-stage optimization problem, which is then transformed into Markov decision processes. A model-free framework applying two state-of-the-art deep reinforcement learning algorithms is proposed to solve the optimization problem without requiring complex environmental modeling and prior knowledge. Numerical results demonstrate that the proposed framework outperforms benchmark algorithms regarding both consumer satisfaction preservation and carbon emission reduction.

Accurate detection of surface defects on railway tracks is critical for safe railway operation. Most existing models rely solely on Red–Green–Blue (RGB) images, limiting their ability to capture structural information. Incorporating depth features provides richer spatial cues, significantly improving detection accuracy. However, current Red–Green–Blue and Depth (RGB-D) dual-stream models suffer from high computational complexity and hardware dependencies, making them impractical for real-world deployment. To address these limitations, we propose DAHNet, an asymmetric knowledge distillation model with a teacher–student architecture. DAHNet-T serves as the teacher network, taking RGB-D inputs and integrating a cross-modal attention feature enhancement (CAFE) module to capture contextual information, along with a depth feature interaction block (DFIB) for efficient cross-modal fusion. DAHNet-S is the student network, a lightweight single-stream RGB model employing depthwise separable convolutions to reduce computation. We introduce a multi-level distillation strategy with dynamic temperature scaling to balance coarse-grained and fine-grained knowledge transfer, while incorporating contrastive learning and structural loss to improve pixel-level accuracy. Extensive experiments on the NEU RSDDS-AUG dataset demonstrate that our distilled model DAHNet-KD outperforms state-of-the-art methods. Compared to DAHNet-T, the number of parameters is reduced from 87.72 MParams to 13.97 MParams, and the computational cost decreases from 19.79 GFLOPs to 5.41 GFLOPs. The proposed model achieves superior performance across various evaluation metrics and also generalizes well on other public datasets. Therefore, the model provides a lightweight and high-accuracy solution for deployment on mobile devices in real-world industrial scenarios.

Wireless channels are inherently vulnerable to malicious jamming. Recently, end-to-end (E2E) learning has emerged as a promising paradigm for intelligent physical-layer design, particularly in multiple-input multiple-output (MIMO) load modulation array (LMA) systems, due to its channel-adaptive capability. By treating jamming as part of the channel state, we investigate robust E2E communication under adversarial conditions. However, two major challenges hinder practical deployment: (1) the high training and communication overhead caused by frequent retraining under dynamic channels and jamming, and (2) the unrealistic reliance on perfect channel state information (CSI) and feedback in resource-constrained systems. To address these issues, we propose Channel-Adaptive Hierarchical End-to-End learning (CAHE), a framework that reformulates E2E learning under stochastic channels as a dynamic mapping from channel realizations to optimal network parameters. CAHE consists of a channel-adaptive generator layer that dynamically produces encoder and decoder parameters, and an E2E learning layer that deploys these parameters for adaptive transceiver design. To reduce CSI feedback, we introduce a compact channel embedding mechanism for efficient representation. For jamming scenarios, we further propose a differential block-averaged channel estimation method to enable accurate anti-jamming channel estimation. In addition, we design a jamming-resilient network that embeds convolutional neural network (CNN)-learned jamming features into the decoder generator, enhancing perception of the communication state and improving robustness to random and dynamic jamming. Extensive analysis demonstrates the interpretability and feasibility of CAHE, and simulations confirm its significant performance gains over existing schemes.

Accurate prediction of drug-target interactions (DTIs) is crucial for accelerating drug discovery and reducing experimental costs. However, challenges such as sparse interactions and heterogeneous datasets complicate this prediction. In this study, we hypothesize that leveraging nonnegative matrix completion and integrating heterogeneous similarity information from multiple biological views can improve the accuracy, interpretability, and scalability of DTI prediction. To validate this, we propose two multi-view fused nonnegative matrix completion methods that combine three key components: (1) a nonnegative matrix completion framework that avoids heuristic rank selection and ensures biologically interpretable predictions; (2) a linear multi-view fusion mechanism, where weights over multiple drug and target similarity matrices are jointly learned through linearly constrained quadratic programming; and (3) multi-graph Laplacian regularization to preserve structural properties within each view. The optimization is performed using two efficient proximal linearization-incorporated block coordinate descent algorithms. Extensive experiments on four gold-standard datasets and a larger real-world dataset demonstrate that our models consistently outperform state-of-the-art single-view, multi-view and deep learning-based DTI prediction methods. Furthermore, ablation studies confirm the contribution of each model component, and scalability analysis highlights the computational efficiency of our approach.

With the proliferation of mobile sensing technology, substantial time series data have been generated and accumulated in various distributed domains, providing the basis for practical applications. Federated Learning (FL) has emerged as an essential framework for machine learning on decentralized data, especially with its potential for privacy-preserving. However, deployed with distributed and various edge devices, existing FL frameworks struggle to address the statistical heterogeneity of spatial feature distribution shifts from heterogeneous sensors. Due to the complex temporal dynamics of real-world time series data, temporal feature heterogeneity may result in suboptimal model adaptation performance. To this end, we propose a novelFederated learning approach withOrthogonal regularization forSpatial-Temporal heterogeneity (FedOST) on time series classification. It is featured in three aspects: (1) In the local training phase, we utilize an orthogonal projection to disentangle and align the shared and personalized features, as well as complementary information from different views of the time series data to formulate a robust multi-view training. (2) In the global aggregation phase, we adopt trainable global prototypes to improve feature space separation through orthogonal constraint, to serve as refined global knowledge in the local training. (3) In the testing phase, we leverage an uncertainty-aware test-time adaptation scheme to tackle the temporal feature shifts of unlabeled test data. We conduct extensive evaluations on real-world datasets, where FedOST outperforms existing state-of-the-art baselines with significant advantages.

Graph anomaly detection (GAD) refers to identifying abnormal graph nodes or edges that heavily deviate from normal observations. Existing approaches inevitably suffer from the influence of imbalanced data and privacy protection. This shortcoming poses challenges in optimizing node embeddings and detecting multitype anomalies simultaneously, resulting in decreased accuracy of existing GAD models. To address this shortcoming, we introduce a new federated learning model for graph anomaly detection (FedGAD). FedGAD enables collaborative unsupervised learning among decentralized data centers without requiring direct access to the distributed subgraphs. Specifically, FedGAD masks and reconstructs the neighborhood features to enhance the knowledge of node representations. Considering the data diversity across distributed clients, we also design a cross-clients' node representation module that enables nodes to reconstruct neighbors by leveraging information from other clients. Furthermore, we use a multiscale contrastive learning function, which includes both structure-level and contextual-level learning functions, to detect graph anomalies in the condition that subgraphs located at different clients show imbalanced data distributions. Experimental results on seven benchmark datasets demonstrate the superior performance of FedGAD compared with baseline methods, verifying its capability of improving GAD performance.

Personalized federated learning (PFL) on graphs is an emerging field focusing on the collaborative development of architectures across multiple clients, each with distinct graph data distributions while adhering to strict privacy standards. This area often requires extensive expert intervention in model design, which is a significant limitation. Recent advancements have aimed to automate the search for graph neural network architectures, incorporating large language models (LLMs) for their advanced reasoning and self-reflection capabilities. However, two technical challenges persist. First, although LLMs are effective in natural language processing, their ability to meet the complex demands of graph neural architecture search (GNAS) is still being explored. Second, while LLMs can guide the architecture search process, they do not directly solve the issue of client drift due to heterogeneous data distributions. To address these challenges, we introduce a novel method, Personalized Federated Graph Neural Architecture Search (PFGNAS). This approach employs a task-specific prompt to identify and integrate optimal GNN architectures continuously. To counteract client drift, PFGNAS utilizes a weight-sharing strategy of supernet, which optimizes the local architectures while ensuring client-specific personalization. Extensive evaluations show that PFGNAS significantly outperforms traditional PFL methods, highlighting the advantages of integrating LLMs into personalized federated learning environments.

Despite the rapid development of Chinese vision-language models (VLMs), most existing Chinese vision-language (VL) datasets are constructed on Western-centric images from existing English VL datasets. The cultural bias in the images makes these datasets unsuitable for evaluating VLMs in Chinese culture. To remedy this issue, we present a new Chinese Vision-Language Understanding Evaluation (CVLUE) benchmark dataset, where the selection of object categories and images is entirely driven by Chinese native speakers, ensuring that the source images are representative of Chinese culture. The benchmark contains four distinct VL tasks ranging from image-text retrieval to visual question answering, visual grounding and visual dialogue. We present a detailed statistical analysis of CVLUE and provide a baseline performance analysis with several open-source multilingual VLMs on CVLUE and its English counterparts to reveal their performance gap between English and Chinese. Our in-depth category-level analysis reveals a lack of Chinese cultural knowledge in existing VLMs. We also find that fine-tuning on Chinese culture-related VL datasets effectively enhances VLMs' understanding of Chinese culture.

In the drug discovery process, the low success rate of drug candidate screening often leads to insufficient labeled data, causing the few-shot learning problem in molecular property prediction. Existing methods for few-shot molecular property prediction overlook the sample selection bias, which arises from non-random sample selection in chemical experiments. This bias in data representativeness leads to suboptimal performance. To overcome this challenge, we present a novel method named Contextual Representation Anchor Network (CRANet), where an anchor refers to a cluster center of the representations of molecules and serves as a bridge to transfer enriched contextual knowledge into molecular representations and enhance their expressiveness. CRANet introduces a dual-augmentation mechanism that includes context augmentation, which dynamically retrieves analogous unlabeled molecules and captures their task-specific contextual knowledge to enhance the anchors, and anchor augmentation, which leverages the anchors to augment the molecular representations. We evaluate our approach using the MoleculeNet and FS-Mol benchmarks, as well as through domain transfer experiments. The outcomes indicate that CRANet surpasses current state-of-the-art methods by 0.10% to 5.48% in AUC and 2.52% in ΔAUC-PR metrics, showcasing its exceptional generalization abilities.

Supervised Fine-tuning has been pivotal in training autoregressive language models, yet it introduces exposure bias. To mitigate this, Post Fine-tuning, including on-policy and off-policy methods, has emerged as a solution to enhance models further. However, each has its limitations regarding performance enhancements and susceptibility to overfitting. In this paper, we introduce a novel on-policy approach called Evolution Strategy Optimization (ESO), which is designed by harnessing the principle of biological evolution, namely survival of the fittest. Particularly, we consider model tuning as an evolution process, and each output sentence generated by the model can provide a perturbation signal to the model parameter space. Then, the fitness of perturbation signals is quantified by the difference between its score and the averaged one offered by a reward function, which guides the optimization process. Empirically, the proposed method can achieve superior performance in various tasks and comparable performance in the human alignment task.

End-to-end (E2E) learning has emerged as a key technology for intelligent physical-layer communications, demonstrating significant potential in multiple-input multiple-output (MIMO) load-modulation array (LMA) systems. However, practical deployment faces two major challenges: high online training overhead due to retraining on new channel realizations and impractical reliance on perfect channel state information (CSI) and feedback in resource-constrained systems. To address these issues, we propose Channel-Adaptive Hierarchical End-to-End Learning (CAHE), a novel framework that reformulates E2E learning under stochastic channels as a dynamic mapping problem between channel realizations and optimal E2E network parameters. CAHE comprises a channel-adaptive generator layer, which dynamically generates encoder and decoder parameters tailored to each channel, and an E2E learning layer, which deploys these parameters for adaptive encoding and decoding. Moreover, to reduce channel feedback overhead, we introduce an efficient channel embedding mechanism, achieving up to 75% lossless compression of channel data. Furthermore, we develop a pilot-assisted alternative scheme to address scenarios where perfect CSI is unavailable. Simulation results demonstrate that CAHE significantly outperforms existing methods, achieving up to 9 dB gain at a bit error rate of 3×10−5, while the pilot-based scheme exhibits robust performance in imperfect CSI scenarios.

Heterogeneous graphs are increasingly used to represent complex real-world scenarios with diverse entities and interactions by meta structures. Recently, the search of meta structures is combined with graph neural architecture search to automatically extract the semantic knowledge for various tasks in heterogeneous graphs. However, prior research primarily focuses on identifying meta structures that are universally applicable across all nodes in a graph, neglecting the variations in meta structure selection that arise from the unique features and topology of individual nodes. To address this challenge, we introduce a Node-Centric approach to search Meta Structures in heterogeneous graphs (NC-MS for short). NC-MS implements a node level method that discover meaningful meta structures tailored to each node, capturing subtle differences in meta structure choices between nodes and providing nuanced identification. Additionally, NC-MS utilizes an efficient and differentiable network to enhance operational efficiency. Empirical studies across three real-world datasets validate the superiority of NC-MS, demonstrating its ability to outperform existing models in heterogeneous graph neural networks.

The reconfigurable intelligent surface (RIS)-based carrier modulation technology, which operates without conventional radio-frequency chains, offers hardware simplicity and cost efficiency. In particular, most current RIS-based amplitude modulation schemes rely on the ON/OFF state switching or amplifier gain control of RIS elements, and ambiguity-free high-order phase modulation requires correspondingly highresolution phase shifts. However, practical hardware limitations often impose a low-resolution phase shift and near-constant reflecting amplitude of RIS elements, complicating high-order modulation. To address this, we propose an efficient composite phase shift keying (CPSK) scheme to generate virtual highorder rectangular quadrature amplitude modulation (RQAM) signals at the receiver, referred to as CPSK-RQAM. By decomposing high-order RQAM signals into multiple low-order quadrature phase shift keying signal components, with each component transmitted by a distinct RIS group, CPSK-RQAM eliminates the need for additional amplitude control and ensures robustness to phase-shift quantization errors. Unlike the similar schemes based on conventional I/Q decomposition and equal grouping, CPSK-RQAM achieves a significantly larger minimum Euclidean distance by employing a tailored grouping criterion. Furthermore, we derive the closed-form expressions for the approximate symbol error probability (SEP) of CPSK-RQAM under maximum likelihood detection. Simulation results validate the theoretical analysis and demonstrate the superiority of CPSKRQAM over the state-of-the-art schemes.

Denoising Diffusion Probabilistic Models (DDPMs) have recently demonstrated exceptional performance in generating high-quality data. In this work, we propose CommunityDF, a novel framework that applies DDPMs to the community search problem, which involves identifying subgraphs containing nodes closely related to a given query node. However, three key challenges arise in this context: (I) learning effective node representations from limited examples, (II) discretizing continuous node representations into community members, and (III) reducing the number of diffusion steps without sacrificing performance. To tackle these, CommunityDF introduces several innovations. First, we focus on subgraphs around the query node to reduce interference from unrelated nodes, improve scalability. We then employ a contrastive learning approach, treating node states at different diffusion steps as positive examples and designing various negative sampling strategies to learn high-quality node representations from limited examples. Second, we propose a dynamic thresholding mechanism that effectively converts continuous representations into community members. Finally, we reduce the number of diffusion steps by leveraging the rough communities to initialize the process with rough community structures, which accelerates convergence while maintaining high accuracy. Extensive experiments on seven real-world datasets demonstrate that CommunityDF outperforms existing methods by 16%-47%, establishing it as a state-of-the-art solution for community search. The source code is available at https://github.com/JiazunChen/CommunityDF.

Drug discovery is crucial for identifying candidate drugs for various diseases. However, its low success rate often results in a scarcity of annotations, posing a few-shot learning problem. Existing methods primarily focus on single-scale features, overlooking the hierarchical molecular structures that determine different molecular properties. To address these issues, we introduce Universal Matching Networks (UniMatch), a dual matching framework that integrates explicit hierarchical molecular matching with implicit task-level matching via meta- learning, bridging multi-level molecular representations and task-level generalization. Specifically, our approach explicitly captures structural features across multiple levels—atoms, substructures, and molecules—via hierarchical pooling and matching, facilitating precise molecular representation and comparison. Additionally, we employ a meta-learning strategy for implicit task-level matching, allowing the model to capture shared patterns across tasks and quickly adapt to new ones. This unified matching framework ensures effective molecular alignment while leveraging shared meta-knowledge for fast adaptation. Our experimental results demonstrate that UniMatch outperforms state-of-the-art methods on the MoleculeNet and FS-Mol benchmarks, achieving improvements of 2.87% in AUROC and 6.52% in ∆AUPRC. UniMatch also shows excellent generalization ability on the Meta-MolNet benchmark.

Time series anomaly detection aims to identify unusual patterns in data or deviations from systems' expected behavior. The reconstruction-based methods are the mainstream in this task, which learn point-wise representation via unsupervised learning. However, the unlabeled anomaly points in training data may cause these reconstruction-based methods to learn and reconstruct anomalous data, resulting in the challenge of capturing normal patterns. In this paper, we propose a time series anomaly detection method based on implicit neural representation (INR) reconstruction, named TSINR, to address this challenge. Due to the property of spectral bias, TSINR enables prioritizing low-frequency signals and exhibiting poorer performance on high-frequency abnormal data. Specifically, we adopt INR to parameterize time series data as a continuous function and employ a transformer-based architecture to predict the INR of given data. As a result, the proposed TSINR method achieves the advantage of capturing the temporal continuity and thus is more sensitive to discontinuous anomaly data. In addition, we further design a novel form of INR continuous function to learn inter- and intra-channel information, and leverage a pre-trained large language model to amplify the intense fluctuations in anomalies. Extensive experiments demonstrate that TSINR achieves superior overall performance on both univariate and multivariate time series anomaly detection benchmarks compared to other state-of-the-art reconstruction-based methods. Our codes are available here.

Effective molecular representation learning is crucial for advancing molecular property prediction and drug design. Mainstream molecular representation learning methods rely on Graph Neural Networks (GNNs) to model complex molecular structures. However, these GNN-based approaches struggle with three significant issues: insufficient annotations, molecular diversity, and architectural limitations such as over-squashing, which leads to the loss of critical structural details. To address these issues, we introduce a new class of GNNs that integrates the Kolmogorov-Arnold Networks (KANs), known for their robust data-fitting capabilities and high accuracy in small-scale AI + Science tasks. By incorporating KANs into GNNs, our model enhances the representation of molecular structures. We further advance this approach with a variant called SwallowKAN (SKAN), which employs adaptive Radial Basis Functions (RBFs) as the core of the non-linear neurons. This innovation improves both computational efficiency and adaptability to diverse molecular structures. Building on the strengths of SKAN, we propose a new class of GNNs, GNN-SKAN, and its augmented variant, GNN-SKAN+, which incorporates a SKAN-based classifier to further boost the performance. Experiments across 6 classification datasets, 6 regression datasets, and 4 few-shot learning datasets demonstrate that our approach achieves new state-of-the-art performance in terms of accuracy and computational cost.

In recent years, there has been a surge in research on dynamic graph representation learning, primarily focusing on modeling the evolution of temporal-spatial patterns in real-world applications. However, within the domain of discrete-time dynamic graphs, the exploration of temporal edges remains underexplored. Existing approaches often rely on additional sequential models to capture dynamics, leading to high computational and memory costs, particularly for large-scale graphs. To address this limitation, we propose the Input Snapshots Fusion based Dynamic Graph Neural Network (SFDyG), which combines Hawkes processes with graph neural networks to capture temporal and structural patterns in dynamic graphs effectively. By fusing multiple snapshots into a single temporal graph, SFDyG decouples computational complexity from the number of snapshots, enabling efficient full-batch and mini-batch training. Experimental evaluations on eight diverse dynamic graph datasets for future link prediction tasks demonstrate that SFDyG consistently outperforms existing methods.

Parallel and distributed learning frameworks are now widely employed for the training of large language models. However, the extremely large model size and data size have made the communication a bottleneck, severely reducing the training efficiency. In this paper, we introduce two novel hierarchical allreduce algorithms designed for 2D-torus topology. By exploiting the cooperation of the intra-cluster communication link and the inter-cluster communication link, the congestion bottleneck caused by the imbalanced link bandwidth can be alleviated. Numerical results are presented to demonstrate the performance of the proposed algorithms in comparison with existing baseline algorithms.

The supervision of novel psychoactive substances (NPSs) is a global problem, and the regulation of NPSs was heavily relied on identifying structural matches in established NPSs databases. However, violators could circumvent legal oversight by altering the side chain structure of recognized NPSs and the existing methods cannot overcome the inaccuracy and lag of supervision. In this study, we propose a scaffold and transformer-based NPS generation and Screening (STNGS) framework to systematically identify and evaluate potential NPSs. A scaffold-based generative model and a rank function with four parts are contained by our framework. Our generative model shows excellent performance in the design and optimization of general molecules and NPS-like molecules by chemical space analysis and property distribution analysis. The rank function includes synthetic accessibility score and frequency score, as well as confidence score and affinity score evaluated by a neural network, which enables the precise positioning of potential NPSs. Applied STNGS framework with molecular docking and a G protein-coupled receptor (GPCR) activation-based sensor (GRAB), we successfully identify three novel synthetic cannabinoids with activity. STNGS constrains the chemical space to generate NPS-like molecules database with diversity and novelty, which assists in the ex-ante regulation of NPSs.

Recent advances in the design of convolutional neural networks have shown that performance can be enhanced by improving the ability to represent multi-scale features. However, most existing methods either focus on designing more sophisticated attention modules, which leads to higher computational costs, or fail to effectively establish long-range channel dependencies, or neglect the extraction and utilization of structural information. This work introduces a novel module, the Multi-Branch Concatenation (MBC), designed to process input tensors and extract multi-scale feature maps. The MBC module introduces new degrees of freedom (DoF) in the design of attention networks by allowing for flexible adjustments to the types of transformation operators and the number of branches. This study considers two key transformation operators: multiplexing and splitting, both of which facilitate a more granular representation of multi-scale features and enhance the receptive field range. By integrating the MBC with an attention module, a Multi-Branch Attention (MBA) module is developed to capture channel-wise interactions within feature maps, thereby establishing long-range channel dependencies. Replacing the 3x3 convolutions in the bottleneck blocks of ResNet with the proposed MBA yields a new block, the Efficient Multi-Branch Attention (EMBA), which can be seamlessly integrated into state-of-the-art backbone CNN models. Furthermore, a new backbone network, named EMBANet, is constructed by stacking EMBA blocks. The proposed EMBANet has been thoroughly evaluated across various computer vision tasks, including classification, detection, and segmentation, consistently demonstrating superior performance compared to popular backbones.

With the deployment of large-scale antenna arrays, the already limited time-frequency resources are becoming increasingly scarce. In this study, we propose a novel Laplacian Pyramid Channel Completion Network (LPCCNet) designed for channel completion, thereby reducing the demand for time-frequency resources in massive MIMO systems. Compared with existing network models, the proposed LPCCNet, by employing a progressive upsampling architecture, effectively mitigates aliasing effects, suppresses error propagation, and achieves a substantial reduction in computational complexity. The simulation results show that LPCCNet achieves a superior channel completion quality compared to existing methods, particularly in rapidly time-varying scenarios.

Graph neural networks (GNNs) are a prominent trend in graph-based deep learning, known for their capacity to produce high-quality node embeddings. However, the existing GNN framework design is only implemented from the algorithm level, and the hardware architecture of the GPU is not fully utilized.To this end, we propose DCGG, a dynamic runtime adaptive framework, which can accelerate various GNN workloads on GPU platforms. DCGG has carried out deeper optimization work mainly in terms of load balancing and software and hardware matching. Accordingly, three optimization strategies are proposed. First, we propose dynamic 2D workload management methods and perform customized optimization based on it, effectively reducing additional memory operations. Second, a new slicing strategy is adopted, combined with hardware features, to effectively improve the efficiency of data reuse. Third, DCGG uses the Quantitative Dimension Parallel Strategy to optimize dimensions and parallel methods, greatly improving load balance and data locality. Extensive experiments demonstrate that DCGG outperforms the state-of-the-art GNN computing frameworks, such as Deep Graph Library (up to 3.10× faster) and GNNAdvisor (up to 2.80× faster), on mainstream GNN architectures across various datasets.

Federated Learning with Cloud-Edge Collaboration (FL-CEC) has emerged as a cutting-edge paradigm in distributed learning. Efficient resource investment incentive mechanisms are crucial to encouraging clients in FL-CEC to contribute the necessary data and computational resources for training. However, existing studies are inadequate in meeting the incentive design requirements under multi-level information-sharing scenarios. Moreover, current works often rely on specific functional relationships between resource investment and global model accuracy. To bridge these gaps, this paper investigates the incentive problem for data and computational resource investment under multi-level information-sharing levels. We design a resource investment incentive mechanism based on a weighted potential game without depending on any specific functional relationship between data investment and model accuracy. Furthermore, we propose four algorithms to solve resource investment strategies for different levels of information sharing. The complexity and convergence rates of the proposed algorithms are thoroughly analyzed. Finally, we construct a simulation incentive platform on the Aliyun. Extensive evaluations demonstrate that the proposed scheme effectively enhances social welfare, and improves collaborative training accuracy and efficiency.

Cloud-Edge Collaborative Architecture (CEA) is a prominent framework that provides low-latency and energy-efficient solutions for video stream processing. In Cloud-Edge Collaborative Video Streaming Systems (CEAVS), efficient online offloading strategies for video tasks are crucial for enhancing user experience. However, most existing works overlook budget constraints, which limits their applicability in real-world scenarios constrained by finite resources. Moreover, they fail to adequately address the heterogeneity of video task redundancies, leading to suboptimal utilization of CEAVS's limited resources. To bridge these gaps, we propose an Efficient Online Computing framework for CEAVS (EOCA) that jointly optimizes accuracy, energy consumption, and latency performance through adaptive online offloading and redundancy compression, without requiring future task information. Technically, we formulate computing offloading and adaptive compression under budget constraints as a stochastic optimization problem that maximizes system satisfaction, defined as a weighted combination of accuracy, latency, and energy performance. We employ Lyapunov optimization to decouple the long-term budget constraint. We prove that the decoupled problem is a generalized ordinal potential game and propose algorithms based on generalized Benders decomposition (GBD) and the best response to obtain Nash equilibrium strategies for computing offloading and task compression. Finally, we analyze EOCA's performance bound, convergence rate, and worst-case performance guarantees. Evaluations demonstrate that EOCA effectively improves satisfaction while effectively balancing satisfaction and computational overhead.

To protect both local gradients and estimated parameters in distributed learning, this paper introduces a masked diffusion (MD) strategy, leading to two algorithms: the MD stochastic gradient (MD-SG) and the MD primal-dual stochastic gradient (MPD-SG). The two algorithms distinguish themselves from existing privacy diffusion methods by incorporating two mechanisms: non-zero mean protection noise and a random matrix step-size. The first mechanism ensures the confidentiality of the transmitted values, while the second protects the gradient information. We analyze the mean-square stability and privacy of the proposed methods under standard assumptions. The results indicate that the MPD-SG algorithm, with a sufficiently small parameter γ, can achieve better steady-state performance than the MD-SG algorithm in heterogeneous data scenarios. Finally, simulations illustrate the effectiveness of the proposed algorithms and support the theoretical analysis.

With the explosion of connected devices and Internet-of-Things (IoT) services in the smart city, the challenge to meet the demands of urban computing is increasingly prominent. Recent advances in vehicle-to-everything (V2X) communications promote urban Internet-connected vehicles to become excellent candidates for computing tasks. However, due to the limited computing capacity of vehicles, conducting computation in the vehicular networks themselves is insufficient to satisfy the demands of smart city applications. Edge computing, which delivers computing tasks to edge servers (e.g., base stations, BSs, or roadside units, RSUs) with plenty of computing resources, could be a possible solution. However, the static deployment of edge servers may cause severe load unbalance among servers in both real-time communication and computation, thereby decreasing the system performance. This paper explores a two-hop vehicle-assisted edge computing network framework in which vehicles are able to offload the tasks beyond their capabilities to underloaded edge servers relaying via neighbor vehicles. According to the state of the time-varying vehicular environment and the dynamic traffic loads among RSUs, we formulate the task offloading, relay node selection, and resources allocations problem as a Markov decision process (MDP) aiming at maximizing the performance of the computation offloading capacity with the considerations of load balancing and latency constraints. We propose a deep reinforcement learning (DRL) algorithm with a DNN as Q action-value function approximator to solve this problem. Extensive simulation results reveal that the proposed scheme can significantly improve the system performance compared to other state-of-the-art algorithms.

With the rapid development of mobile network technology and wearable mobile devices, user-scenario interactions generate a large amount of user behavioral data in the form of multivariate event sequences. Due to data isolation, these multi-scenario events need to be jointly trained to achieve better prediction results. However, traditional federated learning methods face significant challenges when handling distributed event sequences. And the effectiveness of existing modeling approaches for event sequences in federated contexts has not been thoroughly explored. To this end, we propose Federated Multivariate Temporal Point Processes (FedMTPP), which enables learning from distributed event sequences within a novel federated learning framework and leverages efficient event modeling technology, MTPP, to forecast future events. Specifically, FedMTPP restores the temporal structure of the original event sequence by rearranging event embeddings and redesigns the autoregressive-based hidden representation computation in traditional MTPP, making it more suitable for federated prediction tasks. Additionally, FedMTPP incorporates advanced encryption techniques to effectively safeguard user privacy and security. Experimental results on both synthetic and real datasets demonstrate that FedMTPP substantially improves the performance of local models and achieves results comparable to state-of-the-art centralized MTPP methods.

The development of artificial intelligence (AI) enables vehicular edge computing (VEC) servers to be able to provide more intelligent services. However, the limited storage resources of VEC servers constrain the deployment of intelligent service contents, which greatly restricts the intelligence level of the VEC network. To resolve this problem, we first design a novel vehicle-assisted VEC network architecture and further propose VaCo, aVehicle-assistedCollaborative caching system. VaCo allows VEC servers to download the cached service content from any vehicle in the VEC network to support task offloading. VaCo mainly considers the real-time scheduling problem of vehicle storage resources under the dynamic VEC network and the benefit problem caused by invoking vehicle resources under the highly dynamic load environment. VaCo models the vehicle storage resources as an independent resource pool and deploys a cross-VEC server content retrieval mechanism to achieve unified and efficient management of the storage resources of the vehicle cluster and the VEC server cluster. Then, we propose a multi-swarm collaborative optimization scheme to jointly optimize the service failure rate and cost, and further propose a Pareto-based optimization scheme to ensuring that VaCo can correctly evaluate the benefits of invoking vehicle resources in a dynamic VEC network. Finally, we implement VaCo and conduct extensive evaluations on real-world dataset. The experimental results on the real trajectory dataset show that VaCo can effectively utilize vehicle resources and ensure the benefits of both vehicles and VEC servers simultaneously.

Personalized federated learning (PFL) enhances the performance of customized client models through collaborative training without compromising data privacy and ownership. Some previous PFL methods rely on rich prior knowledge about the types of data heterogeneity (such as class imbalance or feature skew), which greatly limits their application ranges. In this paper, we study theUniversal Personalization in Federated Learning (UniPFL), the problem that has no prior knowledge about the types of data heterogeneity. In real-world PFL scenarios, UniPFL is potential because the data distributions of clients are usually heterogeneous and unknown to the server, where quantity imbalance, class imbalance, feature skew, or hybrid heterogeneity are possible contingencies. To address UniPFL, we proposeFedFD, a novel framework with local data augmentation and global concept fusion, which is based on the recent advances inthe foundation generative models(e.g., diffusion models, BLIP-2). On the client side, FedFD utilizes a diffusion model to assist local training by generating augmented data samples, and is then efficiently fine-tuned to be personalized. On the server side, we customize the aggregation strategies based on model similarities to learn both personalized models and diverse feature concepts. Extensive experiments show that FedFD reaches the state-of-the-art on (1) CIFAR-10 and CIFAR-100 for class imbalance; (2) DomainNet and Office-10 for feature skew, and (3) hybrid heterogeneity with both class and feature shifts.

Hierarchical federated learning (HFL) is a key technology enabling distributed learning with reduced communication overhead. However, practical HFL systems encounter two major challenges: limited resources and data heterogeneity. In particular, limited resources can result in intolerable system latency, while heterogeneous data across clients can significantly degrade model accuracy and convergence rates. To address these issues and fully leverage the potential of HFL, we propose a novel framework called graph-based joint client and resource orchestration. This framework addresses the challenges of practical networks through joint client clustering and resource allocation. First, we propose a learning process where edge servers employ hypernetworks to achieve edge aggregation. This method can generate personalized client models and extract data distributions without directly exposing data distributions. Then, to characterize the joint effects of limited resources and data heterogeneity, we propose a graph-based modeling method and formulate a joint optimization problem that aims to balance data distributions and minimize latency. Subsequently, we propose a graph neural network-based algorithm to tackle the formulated problem with low-complexity optimization. Numerical results demonstrate significant benefits over existing algorithms in terms of convergence latency, model accuracy, scalability, and adaptability to new distributions.

Edge computing significantly enhanced the growth of edge-assistant video streaming applications. However, challenges such as unpredictable wireless conditions, resource constraints, and task redundancy have intertwined impacts on the overall performance of edge video streaming systems (EVS). Therefore, it is essential to have an integrated framework that addresses resource management, computational offloading, and video task preprocessing. Existing optimization strategies often neglect the simultaneous management of computational offloading, resource allocation, and video task preprocessing, leading to a suboptimal system utility. Moreover, they struggle to handle high-dimensional decision variables. On the other hand, learning-based adaptive schemes fall short in integrating distributed decisions and ensuring the scalability of wireless devices. Additionally, current approaches lack adaptive incentives. To bridge these gaps, we propose a novel framework called AIRA, which is based on improved multi-agent reinforcement learning (MARL) and smart contracts. AIRA manages resources, video compression, and adaptive incentives in a distributed manner. It consists of a MARL-driven cooperative learning algorithm (CLA) and a smart contract-guided adaptive incentive mechanism. Leveraging an actor-critic structure, the CLA enables wireless devices to master strategies for resource allocation, video task compression, and offloading, utilizing historical data. Notably, the CLA incorporates an attention mechanism to select pivotal tuples from the observation-action pairings among different agents, ensuring improved scalability and computational prowess. Evaluations based on real-world trajectories demonstrate that AIRA enables adaptive incentives. Compared to state-of-the-art approaches, CLA effectively enhances the long-term system utility and scalability of EVS.

The adversarial robustness is critical to deep neural networks (DNNs) in deployment. However, the improvement of adversarial robustness often requires compromising with the network size. Existing approaches to addressing this problem mainly focus on the combination of model compression and adversarial training. However, their performance heavily relies on neural architectures, which are typically manual designs with extensive expertise. In this article, we propose a lightweight and robust neural architecture search (LRNAS) method to automatically search for adversarially robust lightweight neural architectures. Specifically, we propose a novel search strategy to quantify contributions of the components in the search space, based on which the beneficial components can be determined. In addition, we further propose an architecture selection method based on a greedy strategy, which can keep the model size while deriving sufficient beneficial components. Owing to these designs in LRNAS, the lightness, the natural accuracy, and the adversarial robustness can be collectively guaranteed to the searched architectures. We conduct extensive experiments on various benchmark datasets against the state of the arts. The experimental results demonstrate that the proposed LRNAS method is superior at finding lightweight neural architectures that are both accurate and adversarially robust under popular adversarial attacks. Moreover, ablation studies are also performed, which reveals the validity of the individual components designed in LRNAS and the component effects in positively deciding the overall performance.

In continual learning (CL), the newly arrived data are often out-of-distribution from the previous ones, causing drastic representation shift (RS) when updating the old model on the new data, leading to catastrophic forgetting. In this work, we propose feature boosting calibration (FBC) to tackle this problem. Specifically, an expanded module is trained to learn all the classes, including the old and new classes, discovering critical features missed by the original/old model. Then, an FBC network (FBCN) is trained to exploit these missed features to calibrate the old representations. As the missed features increase the information needed for distinguishing between the old and new classes, FBCN generates the calibrated ones with more transferable features, thus alleviating the RS. Next, given the limited memory to store samples of the old/learned classes, the data are severely imbalanced between the old and new classes. To cope with this problem, we propose blockwise knowledge distillation (BWKD), which splits the softmax layer into blocks according to class frequency and then distills each block separately, resolving data imbalance effectively. Building upon the two improvements, we propose a two-stage training framework for CL, named CKDF-V2, providing an enhanced version of the cascaded knowledge distillation framework (CKDF). Furthermore, we integrate it with a task-token expansion method to develop a novel approach for CL based on the vision transformer (ViT). Extensive experiments show that both a convolutional neural network (CNN) and ViT-based CKDF-V2 obtain favorable results across multiple CL benchmarks.

The human brain is a highly complex neurological system that has been the subject of continuous exploration by scientists. With the help of modern neuroimaging techniques, there has been significant progress made in brain disorder analysis. There is an increasing interest about utilizing artificial intelligence techniques to improve the efficiency of disorder diagnosis in recent years. However, these methods rely only on neuroimaging data for disorder diagnosis and do not explore the pathogenic mechanism behind the disorder or provide an interpretable result toward the diagnosis decision. Furthermore, the scarcity of medical data limits the performance of existing methods. As the hot application of graph neural networks (GNNs) in molecular graphs and drug discovery due to its strong graph-structured data learning ability, whether GNNs can also play a huge role in the field of brain disorder analysis. Thus, in this work, we innovatively model brain neuroimaging data into graph-structured data and propose knowledge distillation (KD) guided brain subgraph neural networks to extract discriminative subgraphs between patient and healthy brain graphs to explain which brain regions and abnormal functional connectivities cause the disorder. Specifically, we introduce the KD technique to transfer the knowledge of pretrained teacher model to guide brain subgraph neural networks training and alleviate the problem of insufficient training data. And these discriminative subgraphs are conducive to learn better brain graph-level representations for disorder prediction. We conduct abundant experiments on two functional magnetic resonance imaging datasets, i.e., Parkinson's disease (PD) and attention-deficit/hyperactivity disorder (ADHD), and experimental results well demonstrate the superiority of our method over other brain graph analysis methods for disorder prediction accuracy. The interpretable experimental results given by our method are consistent with corresponding medical research, which is encouraging to provide a potential for deeper brain disorder study.

Heterogeneous graph neural networks (GNNs) have been successful in handling heterogeneous graphs. In existing heterogeneous GNNs, meta-path plays an essential role. However, recent work pointed out that a simple homogeneous graph model without a meta-path can also achieve comparable results, which calls into question the necessity of a meta-path. In this article, we first present the intrinsic difference between meta-path-based and meta-path-free models, i.e., how to select neighbors for node aggregation. Then, we propose a novel framework to utilize the rich type of semantic information in heterogeneous graphs comprehensively, namely, hybrid aggregation for heterogeneous GNNs (HAGNNs). The core of HAGNN is to leverage the meta-path neighbors and the directly connected neighbors simultaneously for node aggregations. HAGNN divides the overall aggregation process into two phases: meta-path-based intratype aggregation and meta-path-free intertype aggregation. During the intratype aggregation phase, we propose a new data structure called a fused meta-path graph and perform structural semantic aware aggregation on it. Finally, we combine the embeddings generated by each phase. Compared with existing heterogeneous GNN models, HAGNN can take full advantage of the heterogeneity in heterogeneous graphs. Extensive experimental results on node classification, node clustering, and link prediction tasks show that HAGNN outperforms the existing modes, demonstrating the effectiveness and efficiency of HAGNN.

Traditional deep functional map frameworks are widely used for 3D shape matching; however, many methods fail to adaptively capture the relevant frequency information required for functional map estimation in complex scenarios, leading to poor performance, especially under significant deformations. To address these challenges, we propose a novel unsupervised learning-based framework, Deep Frequency Awareness Functional Maps (DFAFM), specifically designed to tackle diverse shape-matching problems. Our approach introduces the Spectral Filter Operator Preservation constraint, which ensures the preservation of critical frequency information. These constraints promote frequency awareness by learning a set of spectral filters and incorporating them as a loss function to jointly supervise the functional maps, pointwise maps, and spectral filters. The spectral filters are constructed using orthonormal Jacobi polynomials with learnable coefficients, enabling adaptive and efficient frequency representation. Furthermore, we propose a refinement strategy that leverages the learned spectral filters and constraints to enhance the accuracy of the final pointwise map. Extensive experiments conducted on multiple benchmark datasets demonstrate that our method outperforms state-of-the-art approaches, particularly in challenging scenarios involving non-isometric deformations and inconsistent topology.

Depression detection is a challenging and crucial task in psychological illness diagnosis. Utilizing online user posts to predict whether a user suffers from depression seems an effective and promising direction. However, existing methods suffer from either poor interpretability brought by the black-box models or underwhelming performance caused by the completely separate two-stage model structure. To alleviate these limitations, we propose a novel capsule network integrated with contrastive learning for depression detection (DeCapsNet). The highlights of DeCapsNet can be summarized as follows. First, it extracts symptom capsules from user posts by leveraging meticulously designed symptom descriptions, and then distills them into class-indicative depression capsules. The overall workflow is in an explicit hierarchical reasoning manner and can be well interpreted by the Patient Health Questionnaire-9 (PHQ9), which is one of the most widely adopted questionnaires for depression diagnosis. Second, it integrates with contrastive learning, which can facilitate the embeddings from the same class to be pulled closer, while simultaneously pushing the embeddings from different classes apart. In addition, by adopting the end-to-end training strategy, it does not necessitate additional data annotation, and mitigates the potential adverse effects from the upstream task to the downstream task. Extensive experiments on three widely-used datasets show that in both within-dataset and cross-dataset scenarios our proposed method outperforms other strong baselines significantly.

Few-shot and zero-shot text classification aim to recognize samples from novel classes with limited labeled samples or no labeled samples at all. While prevailing methods have shown promising performance via transferring knowledge from seen classes to unseen classes, they are still limited by (1) Inherent dissimilarities among classes make the transformation of features learned from seen classes to unseen classes both difficult and inefficient. (2) Rare labeled novel samples usually cannot provide enough supervision signals to enable the model to adjust from the source distribution to the target distribution, especially for complicated scenarios. To alleviate the above issues, we propose a simple and effective strategy for few-shot and zero-shot text classification. We aim to liberate the model from the confines of seen classes, thereby enabling it to predict unseen categories without the necessity of training on seen classes. Specifically, for mining more related unseen category knowledge, we utilize a large pre-trained language model to generate pseudo novel samples, and select the most representative ones as category anchors. After that, we convert the multi-class classification task into a binary classification task and use the similarities of query-anchor pairs for prediction to fully leverage the limited supervision signals. Extensive experiments on six widely used public datasets show that our proposed method can outperform other strong baselines significantly in few-shot and zero-shot tasks, even without using any seen class samples.

Denoising diffusion models have shown great potential in multiple research areas. Existing diffusion-based generative methods on de novo 3D molecule generation face two major challenges. Since majority heavy atoms in molecules allow connections to multiple atoms through single bonds, solely using pair-wise distance to model molecule geometries is insufficient. Therefore, the first one involves proposing an effective neural network as the denoising kernel that is capable to capture complex multi-body interatomic relationships and learn high-quality features. Due to the discrete nature of graphs, mainstream diffusion-based methods for molecules heavily rely on predefined rules and generate edges in an indirect manner. The second challenge involves accommodating molecule generation to diffusion and accurately predicting the existence of bonds. In our research, we view the iterative way of updating molecule conformations in diffusion process is consistent with molecular dynamics and introduce a novel molecule generation method named Geometric-Facilitated Molecular Diffusion (GFMDiff). For the first challenge, we introduce a Dual-track Transformer Network (DTN) to fully excevate global spatial relationships and learn high quality representations which contribute to accurate predictions of features and geometries. As for the second challenge, we design Geometric-facilitated Loss (GFLoss) which intervenes the formation of bonds during the training period, instead of directly embedding edges into the latent space. Comprehensive experiments on current benchmarks demonstrate the superiority of GFMDiff.

Evaluation of pharmacokinetic properties such as absorption, distribution, metabolism, excretion, and toxicity is essential but costly and time-consuming in the research and development of new drug candidates. This challenge has driven many researchers to use multitask learning for in silico predictions. Given extra information such as SMILES (simplified molecular input line entry system) as input data, existing multitask learning methods can usually have a good performance in predicting pharmacokinetics, which are served as labels/tasks. However, most of these methods are based solely on additional information, overlooking potentially available pharmacokinetics, which can be utilized to further enhance the prediction of unavailable ones. To address this limitation, we propose Label Enhanced MultiTask Learning (LEMTL), a framework to improve the prediction accuracy of unknown pharmacokinetics of existing multitask learning methods by aggregating shared feature representations learned from additional information and from known pharmacokinetics. Experiments on synthetic and real datasets demonstrate significant improvements over the associated multitask learning approaches when incorporated within the LEMTL framework.

Recently, graph neural networks (GNNs) have demonstrated outstanding performance in fundamental tasks such as node classification and link prediction, as well as in specialized domains like recommendation systems, fraud detection, and drug discovery. However, their vulnerability to adversarial attacks raises concerns about their reliability in security-critical areas. To address this issue, researchers are exploring various defense methods, including specific attack countermeasures and certifiable robustness approaches. Nevertheless, these strategies are often effective only against limited attack scenarios, and prevailing certification methods prove inadequate when confronted with injection attacks. In this paper, we propose a method named CERT_UIA to enhance the robustness of GNN models against worst-case attacks, specifically targeting the scenario of Universal node Injection Attacks (UIA), thereby filling a gap in the existing literature on certified robustness in this context. Our approach involves a two-stage attack process that replaces the transformations of the topology and feature spaces with equivalent unified feature transformations, unifying the optimization of worst-case perturbations into a single feature space. Furthermore, we empirically evaluate our method on several benchmark datasets and compare it with existing certified methods.

A pivotal challenge in meta-heuristic optimization is the lack of knowledge inheritance in heuristic rules. For example, the Covariance Matrix Adaptation Evolution Strategy (CMA-ES) requires generating historical data from scratch for its adaptation mechanisms for each new instance, demanding an extensive number of fitness evaluations. This severely limits the practicality of meta-heuristics, especially under restricted evaluation budgets, hindering efficient navigation in high-dimensional spaces. To overcome this, we integrate Proximal Policy Optimization (PPO) with a vanilla Evolution Strategy (ES), forming the novel PPO-ES approach. Distinct from other adaptive ES variants like CMA-ES with cumulative path calculation and covariance matrix adaptation, it leverages PPO's capability for dynamic step-size adjustment. Our method streamlines the optimization process and incorporates a meticulously designed reward system to adeptly navigate the scalability challenge, significantly enhancing adaptability and efficiency of ES. PPO-ES, trained on part of the bbob benchmarks, was tested on these and the rest unseen problems, and further validated on bbob-largescale benchmarks with much higher dimensions. Results show that PPO-ES achieves faster or comparable convergence to CMA-ES. Our code can be found online: https://github.com/burningxt/PPO-ES_GECCO24.

E-commerce platforms heavily rely on the wide range of products they offer. Detecting communities within these products is essential for efficient risk management and personalized recommendations. In this paper, we introduce a novel framework that utilizes unsupervised pre-trained social network computing methods to detect communities in e-commerce products. Our key technical contribution lies in capturing domain-specific knowledge from academic social networks during the pre-training phase, which is then transferred to the fine-tuning process of the unsupervised e-commerce product graph model. The framework begins with both unsupervised and supervised pre-training modules, conducting masked graph modeling and node classification tasks on academic social networks. Following this, the pre-trained graph autoencoder is fine-tuned on new datasets. In the final stage, graph clustering on node embeddings is performed using an ensemble method of fast K-means algorithms. We evaluate the proposed framework on large-scale real-world datasets from Amazon’s e-commerce platform. Results demonstrate that our method significantly outperforms other classical benchmark methods, leading the second-best by over 10% in adjusted Rand index, and discovering more than 500 subgraphs in the large-scale e-commerce network.

Zero-shot node classification aims to predict nodes belonging to novel classes that have not been seen in the training. Existing studies focus on transferring knowledge from seen classes to unseen classes, which have achieved good performance in most cases. However, they do not fully leverage the relationships between nodes and overlook the issue of domain bias, affecting overall performance. In this paper, we propose a novel dependency capture and discriminative feature learning (DCDFL) model for zero-shot node classification. First, we address the issue of underutilized relationships between nodes by designing a relation-aware network to capture potentially long-range dependency, which can effectively utilize the available data information. Next, we introduce a domain-invariant adversarial loss to mitigate domain bias, encouraging the model to learn domain-insensitive feature representations. Furthermore, we refine and adapt the representations by leveraging inter-class separability within the metric space. Finally, extensive experiments on three well-known benchmark datasets demonstrate the superiority of our method over the strong baselines.

New intent discovery aims to identify new intents from unlabeled utterances in the open-world scenario. As the fundamental and challenging problem in dialogue systems, new intent discovery attracts increasing attention but is still under exploration. In this paper, we propose a simple and effective new intent discovery framework with multi-view clustering. Specifically, we first adopt a double-branch representation learning strategy to learn high-quality utterance representations. Then we conduct a multi-view clustering method to obtain the satisfactory cluster assignment in an iterative manner, thus fulfilling the new intent discovery task. Extensive experimental results on three widely-used datasets demonstrate that our proposed method outperforms other strong baselines in most cases.

Recently link weight prediction has attracted an increasing research interest due to its merits in quantifying the strength between nodes within a graph. Nonetheless, current link weight prediction methods focus solely on graph topology, disregarding node feature information embedded in graphs. In real-world applications, we often collect heterogeneous graph data where multiple types of nodes linked by multiple types of edges are available for analysis, and it is essential and challenging to quantify the proximity of different types of nodes. To solve this challenge, we present a new model for Heterogeneous Graph Link Weight Prediction (HLWP for short). In HLWP, message passing in heterogeneous graph neural networks is described as a meta structure, which can be effectively designed by Differentiable Neural Architecture Search (DARTS) algorithms. Thus, HLWP can enhance the message passing in heterogeneous graphs by DARTS. In addition, HLWP employs a perturbation-based algorithm to enhance stability and precision. Through empirical experiments conducted on three real-world datasets, we demonstrate that HLWP achieves accurate predictions of link weights. Our results highlight the superiority of HLWP over existing methods for link weight prediction and baseline GNN models in terms of accurately predicting link weights within heterogeneous graphs.

Due to the characteristics of open links, immense coverage and dynamic network topology, access authentication is crucial and challenging in integrated satellite-terrestrial back-haul networks (ISTBNs). Based on the access behavior of users, this paper proposes an ISTBN access authentication approach to prevent the spoofing attack, which is the most prominent security threat in ISTBNs. Concretely, we model the access authentication as a machine learning classification task with spatial-temporal characteristics. In this way, the dependency on identification information can be avoided, which is easy to be copied by adversaries in spoofing attacks. Then, we design an ensemble learning classifier to judge the legality of the access to ISTBNs. By organically integrating tree-based models and deep learning through the soft voting method, the detection performance of unauthorized access can be improved. Experiment results validate the effectiveness of our approach for preventing unauthorized access to ISTBNs.

The recent prosperity of temporal graph research redefines many traditional concepts on static graphs, such as triangle, motif,$k$-core, etc. Inspired by that, we propose a novel$(k, \delta)$-truss on temporal graphs, which requires its triangles to exist in short enough time windows ever. The$(k,\delta)$-truss satisfies both static and temporal cohesion, while the original$k$-truss is its special case when$\delta=\infty$. In order to address the$(k, \delta)$-truss query, we propose both index-free and index-based approaches. By leveraging the dual containment relation on$(k, \delta)$-trusses, our indexes can compress all$(k, \delta)$-trusses losslessly into map or tree structures with dramatically less space, so that a specific$(k,\ \delta)$-truss can be retrieved from indexes in the optimal time. To enable our index to scale to large temporal graphs, we develop two index construction algorithms that can reduce redundant computation significantly, based on truss decomposition and truss maintenance respectively. The experimental results demonstrate that index-based approaches process queries in interactive time and outperform the index-free approach by 2~4 orders of magnitude, while indexes achieve compression ratios up to 10-4.

In the context of privacy-aware social web services, secure causal reasoning on graphs is an essential task, for example, identifying the key influential nodes/users through which information, ideas, trends spread across the network. Despite the significant advancements achieved by neural network-based methods in these domains, existing research often assumes complete observable of graph data, an assumption that conflicts with privacy concerns in social web services. This assumption may not hold in numerous real-world social analysis applications due to privacy restrictions. In such scenarios, some graph links might remain unobserved until the model is trained, posing a critical challenge for privacy preservation. To address this issue, we introduce a novel model named Causal Effect on Coarsened Graphs (Cecoar), which is specifically designed for privacy-aware social web services. This model employs a new type of coarsened graph that accurately represents partially observed nodes and edges, aligning with the privacy constraints of social web services. We have developed three crucial modules to enable causal estimation on coarsened graphs: (1) a graph sampling module, (2) a confounder encoder, and (3) a graph-based encoder with causal prediction layers. Furthermore, we present a comprehensive theoretical analysis of the proposed model, demonstrating its performance and generalization capabilities. We evaluate our model on two benchmark datasets for causal estimation, and the results confirm its effectiveness in handling graph data containing private information. The code is in https://github.com/liu-yang-maker/Cecoar-ICWS24.

Self-supervised learning (SSL) has become one of the most popular learning paradigms and has achieved remarkable success in the graph field. Recently, a series of pre-training studies on heterogeneous graphs (HGs) using SSL have been proposed considering the heterogeneity of real-world graph data. However, verification of the robustness of heterogeneous graph pre-training is still a research gap. Most existing researches focus on supervised attacks on graphs, which are limited to a specific scenario and will not work when labels are not available. In this paper, we propose a novel unsupervised heterogeneous graph rewriting attack via node clustering (HGAC) that can effectively attack HG pre-training models without using labels. Specifically, a heterogeneous edge rewriting strategy is designed to ensure the rationality and concealment of the attacks. Then, a tailored heterogeneous graph contrastive learning (HGCL) is used as a surrogate model. Moreover, we leverage node clustering results of the clean HGs as the pseudo-labels to guide the optimization of structural attacks. Extensive experiments exhibit powerful attack performances of our HGAC on various downstream tasks (i.e., node classification, node clustering, metapath prediction, and visualization) under poisoning attack and evasion attack.

Recent years have witnessed remarkable advance-ments in entity alignment, which endeavors to identify entities that represent the same real-world objects across different knowledge graphs (KGs). Nonetheless, prevailing approaches predominantly operate within closed-domain scenarios, rendering them inadequate for handling unmatchable entities. To address this challenge, we propose a retrieval augmented large language model framework (RALLM) to leverage the reasoning capacities of large language models (LLMs) to achieve open-set entity alignment, which not only enables the identification of equivalent entities for matchable entities but also addresses the identification of unmatchable ones. Specifically, we propose a novel retrieval augmentation method that leverages both textual and structural information of entities to retrieve potential equivalent candidates. Subsequently, we employ an iterative process to prompt the LLM to discern the equivalence between the retrieved candidate entity and the entity requiring alignment. To mitigate issues related to many-to-one alignment prediction and enhance alignment efficacy, we devise a memory mechanism to store highly confident aligned entity pairs and provide reminders to the LLM when a candidate entity has been matched. Our experimental findings underscore the superior performance of RALLM, highlighting the potential of LLMs in facilitating open-set entity alignment tasks.

Spatiotemporal traffic forecasting plays a critical role in intelligent transportation systems, which empowers diverse urban services. Existing traffic forecasting frameworks usually devise various learning strategies to capture spatiotemporal correlations from the perspective of volume itself. However, we argue that previous traffic predictions are still unreliable due to two aspects. First, the influences of context factor-wise interactions on dynamic region-wise correlations are under exploitation. Second, the dynamics induce the credibility issue of forecasting that has not been well-explored. In this paper, we exploit the informative traffic-related context factors to jointly tackle the dynamic regional heterogeneity and explain the stochasticity, towards a credible uncertainty-aware traffic forecasting. Specifically, to internalize the dynamic contextual influences into learning process, we design a context-cross relational embedding to capture interactions between each context, and generate virtual graph topology to dynamically relate pairwise regions with context embedding. To quantify the prediction credibility, we attribute data-side aleatoric uncertainty to contexts and re-utilize them for aleatoric uncertainty quantification. Then we couple a dual-pipeline learning with the same objective to produce the discrepancy of model outputs and quantify model-side epistemic uncertainty. These two uncertainties are fed through a spatiotemporal network for extracting uncertainty evolution patterns. Finally, comprehensive experiments and model deployments have corroborated the credibility of our framework.

Unsupervised Graph Domain Adaptation (UGDA) has emerged as a practical solution to transfer knowledge from a label-rich source graph to a completely unlabelled target graph. However, most methods require a labelled source graph to provide supervision signals, which might not be accessible in the real-world settings due to regulations and privacy concerns. In this paper, we explore the scenario of source-free unsupervised graph domain adaptation, which tries to address the domain adaptation problem without accessing the labelled source graph. Specifically, we present a novel paradigm called GraphCTA, which performs model adaptation and graph adaptation collaboratively through a series of procedures: (1) conduct model adaptation based on node's neighborhood predictions in target graph considering both local and global information; (2) perform graph adaptation by updating graph structure and node attributes via neighborhood contrastive learning; and (3) the updated graph serves as an input to facilitate the subsequent iteration of model adaptation, thereby establishing a collaborative loop between model adaptation and graph adaptation. Comprehensive experiments are conducted on various public datasets. The experimental results demonstrate that our proposed model outperforms recent source-free baselines by large margins.

MOTIVATION: The prediction of cancer drug response is a challenging subject in modern personalized cancer therapy due to the uncertainty of drug efficacy and the heterogeneity of patients. It has been shown that the characteristics of the drug itself and the genomic characteristics of the patient can greatly influence the results of cancer drug response. Therefore, accurate, efficient, and comprehensive methods for drug feature extraction and genomics integration are crucial to improve the prediction accuracy. RESULTS: Accurate prediction of cancer drug response is vital for guiding the design of anticancer drugs. In this study, we propose an end-to-end deep learning model named DeepAEG which is based on a complete-graph update mode to predict IC50. Specifically, we integrate an edge update mechanism on the basis of a hybrid graph convolutional network to comprehensively learn the potential high-dimensional representation of topological structures in drugs, including atomic characteristics and chemical bond information. Additionally, we present a novel approach for enhancing simplified molecular input line entry specification data by employing sequence recombination to eliminate the defect of single sequence representation of drug molecules. Our extensive experiments show that DeepAEG outperforms other existing methods across multiple evaluation parameters in multiple test sets. Furthermore, we identify several potential anticancer agents, including bortezomib, which has proven to be an effective clinical treatment option. Our results highlight the potential value of DeepAEG in guiding the design of specific cancer treatment regimens.

Ambient backscatter communications (AmBC), which utilizes ambient radio frequency (RF) signal for communications, is a promising technology for Internet-of-Things (IoT) with stringent power-and-cost constraints. To address the issue of jammer attacking a multi-antenna AmBC system with M-ary frequency shift keying (MFSK) modulation, we propose two detectors: Maximum A Posterior probability (MAP) detector and Energy-Threshold determination (ETD) detector, which are considered for the frequency spectrum of the jammer overlaps the passband of filters or not. Simulation results demonstrate that implementing multiple antennas and higher order modulation can enhance the system performance. Moreover, when the frequency spectrum of the low power jammer does not overlap with the passband of the filters at the receiver, the MAP detector performs better, conversely, the ETD detector outperforms the MAP detector.

In the Industrial Internet of Things (IIoT), cross-silo federated learning (CSFL) enables entities, such as manufacturers and suppliers to train global models for optimizing production processes while ensuring data privacy. A well-designed incentive mechanism is essential to persuade clients to contribute data resources. However, existing methodologies overlook the dynamic nature of the training process, where the accuracy of the globally trained model and the client’s data ownership change over time. Furthermore, the majority of previous research assumes a defined functional relationship between the data contribution and the model accuracy, which is infeasible in realistic and dynamic training environments. To address these challenges, we design a novel adaptive mechanism for CSFL that inspires organizations to contribute data resources in a dynamic training environment with the aim of maximizing their long-term payoffs. This mechanism leverages multiagent reinforcement learning (MARL) to ascertain near-optimal data contribution strategies from potential game histories without necessitating private organizational information or a precise accuracy function. Experimental results indicate that our mechanism achieves adaptive incentive in dynamic environments and effectively enhances the long-term payoffs of organizations.

Effective resistance (ER) is a fundamental metric for measuring node similarities in a graph, and it finds applications in various domains including graph clustering, recommendation systems, link prediction, and graph neural networks. The state-of-the-art algorithm for computing effective resistance relies on a landmark technique, which involves selecting a node that is easy to reach by all the other nodes as a landmark. The performance of this technique heavily depends on the chosen landmark node. However, in many real-life graphs, it is not always possible to find an easily reachable landmark node, which can significantly hinder the algorithm's efficiency. To overcome this problem, we propose a novel multiple landmarks technique which involves selecting a set of landmark nodes V l such that the other nodes in the graph can easily reach any one of a landmark node in V l . Specifically, we first propose several new formulas to compute ER with multiple landmarks, utilizing the concept of Schur complement. These new formulas allow us to pre-compute and maintain several small-sized matrices related to V l as a compact index. With this powerful index technique, we demonstrate that both single-pair and single-source ER queries can be efficiently answered using a newly-developed V l -absorbed random walk sampling or V l -absorbed push technique. Comprehensive theoretical analysis shows that all proposed index-based algorithms achieve provable performance guarantees for both single-pair and single-source ER queries. Extensive experiments on 5 real-life datasets demonstrate the high efficiency of our multiple landmarks-based index techniques. For instance, our algorithms, with a 1.5 GB index size, can be up to 4 orders of magnitude faster than the state-of-the-art algorithms while achieving the same accuracy on a large road network.

This paper considers the scenario of noisy inputs and compressive diffusion (for reducing communication load) with noisy links over sensor networks. We first study the implementation of diffusion bias-compensated Bayesian adaptation (DBCBA) for noisy inputs, which outperforms the existing solution. Next, an average-estimate step is applied to lessen the impact of link noise in the full diffusion case, yielding a diffusion average-estimate bias-compensated Bayesian adaptation (DABCBA) algorithm. A Bayes-based adaptation construction step is then presented to reconstruct the compressed diffusion information in the presence of link noise, resulting in a compressive DBCBA (CDBCBA) algorithm whose mean and mean-square behaviors are analyzed and the closed-form expression of the steady-state mean-square deviation is derived. In addition, estimators are devised for the input and output noise variances. The excellent performance of our algorithms is demonstrated via numerical examples while the theoretical calculation aligns closely with the simulation results.

In this article, we explore intercity mobility in urban agglomerations by surveying people traveling across cities based on private car trajectory data. Specifically, we first adopt the statistical analysis method to mine the intercity mobility in terms of various metrics of travel trips, so as to gain a preliminary understanding of intercity mobility in urban agglomeration. Then, we utilize the tensor decomposition method to conduct in-depth study on the intercity mobility pattern from the perspectives of complexity and multidimensionality. We construct a 4-D tensor based on private car trajectory and point-of-interest (POI) datasets and define the functional similarity and geographic adjacency between regions. Finally, we design an alternating proximal gradient (APG)-based method to resolve the core tensor and factor matrix, leading to the fine-grained discovery of intercity mobility patterns on administrative divisions in the urban agglomeration. Extensive experiments are conducted to evaluate the analysis of intercity mobility, using a real-world dataset containing one-year private car trajectories from five cities in the selected urban agglomeration. The experiments show that the proposed method successfully captures 20 intercity mobility patterns, in which the factor matrices retrieve the patterns from different dimensions with core tensors characterizing correlations between patterns in factor matrices. Besides, the extracted intercity mobility patterns not only cover administrative areas with frequent intercity interactions, but also contain areas with less intercity interactions. It validates that the intercity mobility is consistent with the regional functions in urban agglomeration.

Rehearsal methods based on knowledge distillation (KD) have been widely used in continual learning (CL). However, given memory constraints, few exemplars contain limited variations of previously learned tasks, impeding the effectiveness of KD in retaining long-term knowledge. The decision boundaries learned by the typical KD strategy overfit the limited exemplars, leading to “shrunk boundaries" of the old classes. To tackle this problem, we propose a novel KD strategy, called One-to-Many Information Matching method (O2MIM), which generates interpolated data by mixing samples between old and new classes, disentangles the supervision information from them and assigns supervision information to them in favor of the old classes. By doing so, the supervision information from a single exemplar can be matched with multiple information from different interpolated images. Moreover, O2MIM utilizes one trainable parameter to create an adaptive KD loss, thereby facilitating a flexible matching process with the designated supervision information. Consequently, O2MIM exploits the exemplar corset more effectively, expanding the shrunk decision boundaries towards the new classes. Next, to incorporate new classes into our classification model, we apply an effective classification training strategy to train a debiased classifier. Combining it with O2MIM, we propose the method of Expanding the Shrinking Decision Boundaries (ESDB), which simultaneously transfers knowledge from the old model via O2MIM and learns new classes by the classification training strategy. Extensive experiments demonstrate that ESDB achieves state-of-the-art performance on diverse CL benchmarks. We also confirm that O2MIM can be used with various label-mixing methods to improve overall performance in CL. The code is available at: https://github.com/CSTiger77/ESDB.

Many machine learning models are naturally multitask, which may involve regression and classification tasks, in which they can be trained by the multitask network to yield a more generalized model with the aid of correlated features. When these learning models are deployed on Internet-of-Things devices, the computation efficiency and the privacy of the data can pose a significant challenge to developing a federated learning (FL) algorithm for both higher learning performance and better privacy protection. In this article, a new FL framework is proposed for a class of multitask learning problems with hard parameter-sharing model through which the learning tasks are reformulated as a multiobjective optimization problem for better performance. Specifically, the stochastic multiple gradient descent approach and differential privacy are integrated into this FL algorithm for achieving a Pareto optimality that obtains a good tradeoff among different learning tasks while providing data protection. The outstanding performance of this algorithm is demonstrated by the empirical experiments on multiMINIST, the Chinese city parking dataset, and Cityscapes dataset.

Being the most cutting-edge generative methods, diffusion methods have shown great advances in wide generation tasks. Among them, graph generation attracts significant research attention for its broad application in real life. In our survey, we systematically and comprehensively review on diffusion-based graph generative methods. We first make a review on three mainstream paradigms of diffusion methods, which are denoising diffusion probabilistic models, score-based genrative models, and stochastic differential equations. Then we further categorize and introduce the latest applications of diffusion models on graphs. In the end, we point out some limitations of current studies and future directions of future explorations.

Graph pooling technique as the essential component of graph neural networks has gotten increasing attention recently and it aims to learn graph-level representations for the whole graph. Besides, graph pooling is important in graph classification and graph generation tasks. However, current graph pooling methods mainly coarsen a sequence of small-sized graphs to capture hierarchical structures, potentially resulting in the deterioration of the global structure of the original graph and influencing the quality of graph representations. Furthermore, these methods artificially select the number of graph pooling layers for different graph datasets rather than considering each graph individually. In reality, the structure and size differences among graphs necessitate a specific number of graph pooling layers for each graph. In this work, we propose reinforced pooling graph neural networks via adaptive hybrid graph coarsening networks. Specifically, we design a hybrid graph coarsening strategy to coarsen redundant structures of the original graph while retaining the global structure. In addition, we introduce multi-agent reinforcement learning to adaptively perform the graph coarsening process to extract the most representative coarsened graph for each graph, enhancing the quality of graph-level representations. Finally, we design graph-level contrast to improve the preservation of global information in graph-level representations. Extensive experiments with rich baselines on six benchmark datasets show the effectiveness of ReiPool1.

Recently, ChatGPT, a representative large language model (LLM), has gained considerable attention. Due to their powerful emergent abilities, recent LLMs are considered as a possible alternative to structured knowledge bases like knowledge graphs (KGs). However, while LLMs are proficient at learning probabilistic language patterns and engaging in conversations with humans, they, like previous smaller pre-trained language models (PLMs), still have difficulty in recalling facts while generating knowledge-grounded contents. To overcome these limitations, researchers have proposed enhancing data-driven PLMs with knowledge-based KGs to incorporate explicit factual knowledge into PLMs, thus improving their performance in generating texts requiring factual knowledge and providing more informed responses to user queries. This paper reviews the studies on enhancing PLMs with KGs, detailing existing knowledge graph enhanced pre-trained language models (KGPLMs) as well as their applications. Inspired by existing studies on KGPLM, this paper proposes enhancing LLMs with KGs by developing knowledge graph-enhanced large language models (KGLLMs). KGLLM provides a solution to enhance LLMs’ factual reasoning ability, opening up new avenues for LLM research.

Next destination recommendation is a crucial research area for understanding human travel behavior. However, existing studies often overlook the problem of underfitting, which arises due to the limited regularity in users' travel patterns. To tackle this issue, we leverage diverse co-occurrence patterns (CoPs) to discover potential user preferences. These patterns capture intersections with similar spatial and temporal characteristics in users' travels. However, traditional graph neural network (GNN)-based approaches struggle to effectively handle complex spatial-temporal CoPs. To overcome these challenges, we propose a novel framework called DHIN (Dynamic Heterogeneous Information Network). Firstly, to address the problem of underfitting, DHIN constructs intricate CoPs by leveraging abundant features and connection relationships. Additionally, to solve the needs of cold-start users, DHIN generates potential connections by capturing dynamic urban hotspots based on global users' travel trajectories. Moreover, to model dynamic heterogeneous information, DHIN utilizes a hierarchical attention mechanism and integrates a dynamic encoder. The mechanism integrates multi-level attention to learn informative embeddings from heterogeneous attributes and structures, while the dynamic encoder processes dynamic temporal information for updating node representations. Finally, extensive experiments conducted on real-world trajectory data demonstrate the effectiveness of the proposed DHIN model.

Human mobility forecasting is the key to energizing considerable mobile computing services. However, we find that the collective mobility suffers the spatiotemporal heterogeneity issue and therefore leads to inferior performances of conventional homogeneous aggregations. Given two fundamental factors, i.e., data and objectives in machine learning, we propose to counter such heterogeneity by improving data utilization and optimization objectives. 1) From data utilization perspective, we discover that such heterogeneity is inherently induced by mobility-related context factors and thus these factors can be exploited to learn heterogeneous mobility patterns. 2) From the optimization perspective, the dependencies among output elements, which give another prior to learning, can extract heterogeneous correlations within output sequences. Specifically, we propose a novel Context-Directional SpatioTemporal Graph Network (CD-STGNet), which tackles the above-mentioned heterogeneity, for achieving accurate mobility predictions. Firstly, we improve data utilization by inputting the encoded context-wise interactions to a direction field learner, which realizes directional spatial aggregations. Secondly, regarding series learning and optimization objectives, a context-trend highway is designed to enable context-aware temporal learning while two regularization objectives are proposed to keep the correlations among predicted elements consistent with the ground-truth. Experiments demonstrate that CD-STGNet surpasses competitive baselines by 13% to 22% and boosts the interpretability of context-directional learning.

Pedestrian trajectory prediction is a critical technology in the evolution of self-driving cars toward complete artificial intelligence. Over recent years, focusing on the trajectories of pedestrians to model their social interactions has surged with great interest in more accurate trajectory predictions. However, existing methods for modeling pedestrian social interactions rely on pre-defined rules, struggling to capture non-explicit social interactions. In this work, we propose a novel framework named DTGAN, which extends the application of Generative Adversarial Networks (GANs) to graph sequence data, with the primary objective of automatically capturing implicit social interactions and achieving precise predictions of pedestrian trajectory. DTGAN innovatively incorporates random weights within each graph to eliminate the need for pre-defined interaction rules. We further enhance the performance of DTGAN by exploring diverse task loss functions during adversarial training, which yields improvements of 16.7% and 39.3% on metrics ADE and FDE, respectively. The effectiveness and accuracy of our framework are verified on two public datasets. The experimental results show that our proposed DTGAN achieves superior performance and is well able to understand pedestrians' intentions.

Recently, Temporal Graph Neural Networks (TGNNs), as an extension of Graph Neural Networks, have demonstrated remarkable effectiveness in handling dynamic graph data. Distributed TGNN training requires efficiently tackling temporal dependency, which often leads to excessive cross-device communication that generates significant redundant data. However, existing systems are unable to remove the redundancy in data reuse and transfer, and suffer from severe communication overhead in a distributed setting. This work introduces Sven, a co-designed algorithm-system library aimed at accelerating TGNN training on a multi-GPU platform. Exploiting dependency patterns of TGNN models, we develop a redundancy-free graph organization to mitigate redundant data transfer. Additionally, we investigate communication imbalance issues among devices and formulate the graph partitioning problem as minimizing the maximum communication balance cost, which is proved to be an NP-hard problem. We propose an approximation algorithm called Re-FlexBiCut to tackle this problem. Furthermore, we incorporate prefetching, adaptive micro-batch pipelining, and asynchronous pipelining to present a hierarchical pipelining mechanism that mitigates the communication overhead. Sven represents the first comprehensive optimization solution for scaling memory-based TGNN training. Through extensive experiments conducted on a 64-GPU cluster, Sven demonstrates impressive speedup, ranging from 1.9x to 3.5x, compared to State-of-the-Art approaches. Additionally, Sven achieves up to 5.26x higher communication efficiency and reduces communication imbalance by up to 59.2%.

Deep Neural Networks (DNNs) have gained widespread adoption in diverse fields, including image classification, object detection, and natural language processing. However, training large-scale DNN models often encounters significant memory bottlenecks, which ask for efficient management of extensive tensors. Heterogeneous memory system, which combines persistent memory (PM) modules with traditional DRAM, offers an economically viable solution to address tensor management challenges during DNN training. However, existing memory management methods on heterogeneous memory systems often lead to low PM access efficiency, low bandwidth utilization, and incomplete analysis of model characteristics. To overcome these hurdles, we introduce an efficient tensor management approach, DeepTM, tailored for heterogeneous memory to alleviate memory bottlenecks during DNN training. DeepTM employs page-level tensor aggregation to enhance PM read and write performance and executes contiguous page migration to increase memory bandwidth. Through an analysis of tensor access patterns and model characteristics, we quantify the overall performance and transform the performance optimization problem into the framework of Integer Linear Programming. Additionally, we achieve tensor heat recognition by dynamically adjusting the weights of four key tensor characteristics and develop a global optimization strategy using Deep Reinforcement Learning. To validate the efficacy of our approach, we implement and evaluate DeepTM, utilizing the TensorFlow framework running on a PM-based heterogeneous memory system. The experimental results demonstrate that DeepTM achieves performance improvements of up to 36% and 49% compared to the current state-of-the-art memory management strategies AutoTM and Sentinel, respectively. Furthermore, our solution reduces the overhead by 18 times and achieves up to 29% cost reduction compared to AutoTM.

To achieve carbon neutrality, reducing carbon emissions is crucial in dispatching problems in smart grid. Though renewable energy such as wind power has low carbon emissions, it suffers from random generation, which makes the thermal power necessary for a stable supply power system. To reduce carbon emissions, the thermal power plants are transformed into carbon capture power plants, which brings new challenges to economic dispatch algorithms. Besides, there are usually many constraints to keep the security operation of power systems, which incurs a large problem scale and high computational cost. Most existing methods either do not consider reducing carbon emissions, or suffer from high computational costs. In this paper, a framework for the carbon capture plants with wind power to reduce both running costs and carbon emissions is designed to support carbon neutrality. To reduce computational cost, initial-training and fine-tuning are used. A deep neural network is employed to describe the relationship between users' load and the constraints, which provides guides for finding the active constraints. Therefore, the problem scale can be significantly decreased, making the optimal dispatching strategy obtained quickly. The experimental results on real-world data show that the proposed framework can obtain the optimal strategy efficiently.

This paper is focused on multiuser load modulation arrays (MU-LMAs) which are attractive due to their low system complexity and reduced cost for millimeter wave (mmWave) multi-input multi-output (MIMO) systems. The existing precoding algorithm for downlink MU-LMA relies on a sub-array structured (SAS) transmitter which may suffer from decreased degrees of freedom and complex system configuration. Furthermore, a conventional LMA codebook with codewords uniformly distributed on a hypersphere may not be channel-adaptive and may lead to increased signal detection complexity. In this paper, we conceive an MU-LMA system employing a full-array structured (FAS) transmitter and propose two algorithms accordingly. The proposed FAS-based system addresses the SAS structural problems and can support larger numbers of users. For LMA-imposed constant-power downlink precoding, we propose an FAS-based normalized block diagonalization (FAS-NBD) algorithm. However, the forced normalization may result in performance degradation. This degradation, together with the aforementioned codebook design problems, is difficult to solve analytically. This motivates us to propose a Deep Learning-enhanced (FAS-DL-NBD) algorithm for adaptive codebook design and codebook-independent decoding. It is shown that the proposed algorithms are robust to imperfect knowledge of channel state information and yield excellent error performance. Moreover, the FAS-DL-NBD algorithm enables signal detection with low complexity as the number of bits per codeword increases.

Hard-label textual adversarial attack is a challenging task, as only the predicted label information is available, and the text space is discrete and non-differentiable. Relevant research work is still in fancy and just a handful of methods are proposed. However, existing methods suffer from either the high complexity of genetic algorithms or inaccurate gradient estimation, thus are arduous to obtain adversarial examples with high semantic similarity and low perturbation rate under the tight-budget scenario. In this paper, we propose a simple and sweet paradigm for hard-label textual adversarial attack, named SSPAttack. Specifically, SSPAttack first utilizes initialization to generate an adversarial example, and removes unnecessary replacement words to reduce the number of changed words. Then it determines the replacement order and searches for an anchor synonym, thus avoiding going through all the synonyms. Finally, it pushes substitution words towards original words until an appropriate adversarial example is obtained. The core idea of SSPAttack is just swapping words whose mechanism is simple. Experimental results on eight benchmark datasets and two real-world APIs have shown that the performance of SSPAttack is sweet in terms of similarity, perturbation rate and query efficiency.

Distribution estimation has been demonstrated as one of the most effective approaches in dealing with few-shot image classification, as the low-level patterns and underlying representations can be easily transferred across different tasks in computer vision domain. However, directly applying this approach to few-shot text classification is challenging, since leveraging the statistics of known classes with sufficient samples to calibrate the distributions of novel classes may cause negative effects due to serious category difference in text domain. To alleviate this issue, we propose two simple yet effective strategies to estimate the distributions of the novel classes by utilizing unlabeled query samples, thus avoiding the potential negative transfer issue. Specifically, we first assume a class or sample follows the Gaussian distribution, and use the original support set and the nearest few query samples to estimate the corresponding mean and covariance. Then, we augment the labeled samples by sampling from the estimated distribution, which can provide sufficient supervision for training the classification model. Extensive experiments on eight few-shot text classification datasets show that the proposed method outperforms state-of-the-art baselines significantly.

Meta-learning methods have shown to be effective in dealing with cold-start recommendation. However, most previous methods rely on an ideal assumption that there exists a similar data distribution between source and target tasks, which are unsuitable for the scenario that only extremely limited number of new user or item interactions are available. In this paper, we propose to boost meta-learning cold-start recommendation with graph neural network (MeGNN). First, it utilizes the global neighborhood translation learning to obtain consistent potential interactions for all new user and item nodes, which can refine their representations. Second, it employs the local neighborhood translation learning to predict specific potential interactions for each node, thus guaranteeing the personalized requirement. In experiments, we combine MeGNN with two representative meta-learning models MeLU and TaNP. Extensive results on two widely-used datasets show the superiority of MeGNN in four different scenarios.

Cross-silo federated learning (CFL) is a distributed learning paradigm that allows organizations (e.g., financial or medical entities) to train a global model on siloed data. Recent studies on mechanisms designed for CFL, however, rarely jointly consider the potential inter-organizational competition and the lack of credibility between organizations, which may discourage organizational participation. In this paper, we investigate the problem of inter-organizational competition and credibility assurance. We propose a distributed trading mechanism, called$TradeFL$, to incentivize organizations to contribute data and computational resources through mutual trading among organizations. Technically, TradeFL characterizes the competition among organizations and compensates for their damage incurred by competition. TradeFL runs on distributed organizations and provides credibility guarantees for compensation through a customized smart contract11Illustration of the prototype: https://github.com/user10963.. We prove that the interaction among organizations that contribute resources to maximize personal payoffs is a weighted potential game. Then, we propose a centralized algorithm and a distributed algorithm to determine the optimal resource contribution. Simulation results and evaluations based on real-world datasets demonstrate that our scheme achieves higher social welfare, increases the amount of contributed data by up to 64%, and improves the accuracy of the global model by at most 23.2%.

Both Transformer and Graph Neural Networks (GNNs) have been used in learning to rank (LTR), they however adhere to two distinct-yet-complementary problem formulations, i.e., ranking score regression based on query-webpage pairs and link prediction within query-webpage bipartite graphs, respectively. Though it is possible to pre-train GNNs or Transformers on source datasets and fine-tune them subject to sparsely annotated LTR datasets separately, the source-target distribution shifts across the pairs and bipartite graphs domains make it extremely difficult to integrate these diverse models into a single LTR framework at a web-scale. We introduce the novel MPGraf model, which utilizes a modular and capsule-based pre-training approach, aiming to incorporate regression capacities from Transformers and link prediction capabilities of GNNs cohesively. Specifically, rather than simply combining these two modules in stacking or parallelizing architectures, MPGraf proposes a three-step method to (1) construct query-webpage graphs from sparsely annotated query-webpage pairs, (2) pre-train the Transformer and GNN modules using source datasets from two domains, and (3) “surgically” fine-tune the integrated models with pretrained weights while handling the distribution shifts of two domains. Extensive experiments have been done to evaluate the performance of MPGraf using real-world datasets collected from large-scale search engines. The results show that MPGraf can outperform baseline algorithms on several major metrics. Further, we deploy and evaluate MPGraf atop a large-scale search engine with realistic web traffics via A/B tests, where we can still observe significant improvement. MPGraf performs consistently in both offline and online evaluations.

Root cause analysis for multi-dimensional time series from large scale micro-service scenarios aims at identifying the set of anomaly attributes by monitoring operational metrics. The online metrics provide a general indication to investigate these attributes' inter-dependencies and can guide the overall exploration process. However, the problem space for the root cause localization still remains largely challenging due to the combinatorial explosion of possible attribute combinations. This leads researchers and practitioners to (a) assume some prior distributions on the data set; (b) assume some data patterns on the attribute combinations; (c) perform pruning techniques to reduce the search space. Furthermore, state-of-the-art root cause analysis methods are often tied to one or more of these assumptions, which makes it difficult to be robust to general scenarios. In this paper, we conclude the heterogeneity in the data patterns by analyzing several open and industrial datasets. A uniform analytical framework, PatternRCA, is proposed such that it can be aware of the patterns in the metrics while avoiding explicit assumptions about them. We design an offline learning procedure that enables the framework to detect existing data patterns, which then can guide it to do fine-grain exploration in online metrics. Our extensive evaluation results show that PatternRCA outperforms state-of-the-art models with better benchmark results in public datasets. Meanwhile, it can scale to complex root cause analysis tasks on datasets with hybrid patterns in production environments.

Decision-focused learning (DFL) combines both machine learning and combinatorial optimization so as to enhance the quality of decision-making. In general, DFL adds an optimization layer after the neural network and solves a focused combinatorial optimization problem. The optimization layer is usually based on KKT conditions or surrogate functions. However, the optimizer associated with the DFL is inflexible and requires a large amount of expert knowledge. Furthermore, it has been shown that the optimizer exhibits a lack of robustness and differentiability when confronted with complex tasks, especially in the graph domain. To solve this problem, we study a more generic situation by taking the optimization layer as a black-box operator on graphs. Then, we design a mixed zeroth-order optimization to differentiate the layer. Specifically, we combine DFL with graph neural networks and present a decision-focused graph neural network named ZO4Graph based on a mixed zeroth-order optimizer. Empirically, we conduct extensive experiments compared with two-stage models on community detection. Numerical results demonstrate that our proposed framework outperforms its peers.

Anomaly detection in multi-type event sequences is a crucial and challenging problem with important applications in various domains, including cybersecurity, finance and healthcare. Temporal point process has emerged as a powerful technique for modeling event sequences and has gained considerable attention in the field of anomaly detection. However, existing temporal point process approaches are either inapplicable to multi-type event sequence data or incur the loss of valuable information in subsequences associated with specific event types. To this end, we propose a novel Multiple Hypothesis Testing based Anomaly Detection method (MultiAD) to detect anomalous multi-type event sequences. The basic idea of MultiAD is to capture the underlying distribution of normal sequences using a neural multivariate point process, based on which the original hypothesis testing problem can be converted into a multiple hypothesis testing using the multivariate time rescaling theorem. By conducting multiple hypothesis tests on the time-rescaled subsequences, MultiAD makes full use of the valuable information contained within individual subsequences. Moreover, we claim that the existing test statistic ignores the sequential information of inter-event time intervals and propose new statistics to address this shortcoming. Finally, we employ the kernel method to obtain a smooth estimator of the distribution of the proposed statistics under the null hypothesis. This ensures a more accurate and reliable computation of the p-value, providing robust statistical inference. Extensive experimental results demonstrate that MultiAD significantly outperforms the state-of-the-art methods on both synthetic and real-world data.

Image template matching refers to localizing a small query image as opposed to a large reference image map. The query image a.k.a template has to be screened across every equal-sized region in the reference map to perform inner-product at pixel-level and the resulting similarity indicates the template location. Due to the domain heterogeneity between template and reference images, the matching performance degrades under dramatic appearance changes. More severely, the asymmetric matching easily leads to over-fitting by suggesting excessively false positive regions. To these ends, we propose an effective template matching method based on contrastive learning to perform a dense and consistent InfoNCEloss during matching. This can increase the matching at finer details, and thus effectively regularizes network training to prevent over-fitting. Extensive experiments on the synthetic aperture radar (SAR) and optical datasets, i.e., SEN1-2 and OS datasets demonstrate that our proposed method outperforms state-of-the-art methods by a large margin.

Decision-based methods have shown to be effective in black-box adversarial attacks, as they can obtain satisfactory performance and only require to access the final model prediction. Gradient estimation is a critical step in black-box adversarial attacks, as it will directly affect the query efficiency. Recent works have attempted to utilize gradient priors to facilitate score-based methods to obtain better results. However, these gradient priors still suffer from the edge gradient discrepancy issue and the successive iteration gradient direction issue, thus are difficult to simply extend to decision-based methods. In this paper, we propose a novel Decision-based Black-box Attack framework with Gradient Priors (DBA-GP), which seamlessly integrates the data-dependent gradient prior and time-dependent prior into the gradient estimation procedure. First, by leveraging the joint bilateral filter to deal with each random perturbation, DBA-GP can guarantee that the generated perturbations in edge locations are hardly smoothed, i.e., alleviating the edge gradient discrepancy, thus remaining the characteristics of the original image as much as possible. Second, by utilizing a new gradient updating strategy to automatically adjust the successive iteration gradient direction, DBA-GP can accelerate the convergence speed, thus improving the query efficiency. Extensive experiments have demonstrated that the proposed method outperforms other strong baselines significantly.

We study the task of spatio-temporal extrapolation that generates data at target locations from surrounding contexts in a graph. This task is crucial as sensors that collect data are sparsely deployed, resulting in a lack of fine-grained information due to high deployment and maintenance costs. Existing methods either use learning-based models like Neural Networks or statistical approaches like Gaussian Processes for this task. However, the former lacks uncertainty estimates and the latter fails to capture complex spatial and temporal correlations effectively. To address these issues, we propose Spatio-Temporal Graph Neural Processes (STGNP), a neural latent variable model which commands these capabilities simultaneously. Specifically, we first learn deterministic spatio-temporal representations by stacking layers of causal convolutions and cross-set graph neural networks. Then, we learn latent variables for target locations through vertical latent state transitions along layers and obtain extrapolations. Importantly during the transitions, we propose Graph Bayesian Aggregation (GBA), a Bayesian graph aggregator that aggregates contexts considering uncertainties in context data and graph structure. Extensive experiments show that STGNP has desirable properties such as uncertainty estimates and strong learning capabilities, and achieves state-of-the-art results by a clear margin.

Multi-view or even multi-modal data is appealing yet challenging for real-world applications. Detecting anomalies in multi-view data is a prominent recent research topic. However, most of the existing methods 1) are only suitable for two views or type-specific anomalies, 2) suffer from the issue of fusion disentanglement, and 3) do not support online detection after model deployment. To address these challenges, our main ideas in this paper are three-fold: multi-view learning, disentangled representation learning, and generative model. To this end, we propose dPoE, a novel multi-view variational autoencoder model that involves (1) a Product-of-Experts (PoE) layer in tackling multi-view data, (2) a Total Correction (TC) discriminator in disentangling view-common and view-specific representations, and (3) a joint loss function in wrapping up all components. In addition, we devise theoretical information bounds to control both view-common and view-specific representations. Extensive experiments on six real-world datasets demonstrate that the proposed dPoE outperforms baselines markedly.

Black-box hard-label adversarial attack on text is a practical and challenging task, as the text data space is inherently discrete and non-differentiable, and only the predicted label is accessible. Research on this problem is still in the embryonic stage and only a few methods are available. Nevertheless, existing methods rely on the complex heuristic algorithm or unreliable gradient estimation strategy, which probably fall into the local optimum and inevitably consume numerous queries, thus are difficult to craft satisfactory adversarial examples with high semantic similarity and low perturbation rate in a limited query budget. To alleviate above issues, we propose a simple yet effective framework to generate high quality textual adversarial examples under the black-box hard-label attack scenarios, named HQA-Attack. Specifically, after initializing an adversarial example randomly, HQA-attack first constantly substitutes original words back as many as possible, thus shrinking the perturbation rate. Then it leverages the synonym set of the remaining changed words to further optimize the adversarial example with the direction which can improve the semantic similarity and satisfy the adversarial condition simultaneously. In addition, during the optimizing procedure, it searches a transition synonym word for each changed word, thus avoiding traversing the whole synonym set and reducing the query number to some extent. Extensive experimental results on five text classification datasets, three natural language inference datasets and two real-world APIs have shown that the proposed HQA-Attack method outperforms other strong baselines significantly.

Molecular representation learning lays the foundation for drug discovery. However, existing methods suffer from poor out-of-distribution (OOD) generalization, particularly when data for training and testing originate from different environments. To address this issue, we propose a new framework for learning molecular representations that exhibit invariance and robustness against distribution shifts. Specifically, we propose a strategy called ``first-encoding-then-separation'' to identify invariant molecule features in the latent space, which deviates from conventional practices. Prior to the separation step, we introduce a residual vector quantization module that mitigates the over-fitting to training data distributions while preserving the expressivity of encoders. Furthermore, we design a task-agnostic self-supervised learning objective to encourage precise invariance identification, which enables our method widely applicable to a variety of tasks, such as regression and multi-label classification. Extensive experiments on 18 real-world molecular datasets demonstrate that our model achieves stronger generalization against state-of-the-art baselines in the presence of various distribution shifts. Our code is available at https://github.com/HICAI-ZJU/iMoLD.

Graph neural networks (GNNs) are powerful models to handle graph data and can achieve state-of-the-art in many critical tasks including node classification and link prediction. However, existing graph neural networks still face both challenges of over-smoothing and over-squashing based on previous literature. To this end, we propose a new Curvature-based topology-aware Dropout sampling technique named CurvDrop, in which we integrate the Discrete Ricci Curvature into graph neural networks to enable more expressive graph models. Also, this work can improve graph neural networks by quantifying connections in graphs and using structural information such as community structures in graphs. As a result, our method can tackle the both challenges of over-smoothing and over-squashing with theoretical justification. Also, numerous experiments on public datasets show the effectiveness and robustness of our proposed method. The code and data are released in https://github.com/liu-yang-maker/Curvature-based-Dropout.

Deciphering 3D genome conformation is important for understanding gene regulation and cellular function at a spatial level. The recent advances of single cell Hi-C technologies have enabled the profiling of the 3D architecture of DNA within individual cell, which allows us to study the cell-to-cell variability of 3D chromatin organization. Computational approaches are in urgent need to comprehensively analyze the sparse and heterogeneous single cell Hi-C data. Here, we proposed scDEC-Hi-C, a new framework for single cell Hi-C analysis with deep generative neural networks. scDEC-Hi-C outperforms existing methods in terms of single cell Hi-C data clustering and imputation. Moreover, the generative power of scDEC-Hi-C could help unveil the differences of chromatin architecture across cell types. We expect that scDEC-Hi-C could shed light on deepening our understanding of the complex mechanism underlying the formation of chromatin contacts.

Accurately predicting the binding affinity between proteins and ligands is crucial for drug discovery. Recent advances in graph neural networks (GNNs) have made significant progress in learning representations of protein-ligand complexes to estimate binding affinities. To improve the performance of GNNs, there frequently needs to look into protein-ligand complexes from geometric perspectives. While the "off-the-shelf" GNNs could incorporate some basic geometric structures of molecules, such as distances and angles, through modeling the complexes as homophilic graphs, these solutions seldom take into account the higher-level geometric attributes like curvatures and homology, and also heterophilic interactions.To address these limitations, we introduce the Curvature-based Adaptive Graph Neural Network (CurvAGN). This GNN comprises two components: a curvature block and an adaptive attention guided neural block (AGN). The curvature block encodes multiscale curvature informaton, then the AGN, based on an adaptive graph attention mechanism, incorporates geometry structure including angle, distance, and multiscale curvature, long-range molecular interactions, and heterophily of the graph into the protein-ligand complex representation. We demonstrate the superiority of our proposed model through experiments conducted on the PDBbind-V2016 core dataset.

Vehicle-to-vehicle (V2V) computation offloading has emerged as a promising solution to facilitate computing-intensive vehicular task processing, where task vehicles (i.e., TaVs) will be requested to offload computing-intensive tasks to server vehicles (i.e., SeVs) in order to keep task delay low. However, it is challenging for TaVs to obtain the optimal V2V computation offloading decisions (i.e., realizing the minimal task delay) due to the constraints, including: 1) incomplete offloading information; 2) degraded Quality-of-Service (QoS) of SeVs; and 3) privacy leakage risks. In this article, we develop a learning-based V2V computation offloading algorithm enhanced by SeV’s ability & trustfulness awareness to solve these problems. We emphasize that the proposed algorithm learns the offloading performance of candidate SeVs based on history offloading selections, without requiring the complete offloading information in advance. Additionally, both the QoS of SeVs and safe V2V computation offloading are enhanced in the proposed learning-based algorithm. Furthermore, we conduct extensive simulation experiments to validate the proposed algorithm. The results demonstrate that the proposed algorithm reduces the average task delay by 35% and 40%, and at the same time decreases the learning regret by 39% and 41%, compared to the algorithms without SeV’s ability and trustfulness awareness.

In time-varying software-defined wireless sensor networks (SDWSNs) for Internet of Things (IoT) applications, the topology may change due to the interference or abnormal events, thus leading to network performance degradation. In this article, an energy-efficient topology control (TC) mechanism applied for IoT-oriented SDWSNs is proposed to maximize the network energy efficiency (EE) during the dynamic topology maintenance. First, a hierarchical SDWSN architecture consisting of the cluster-based sensing network and the programmable relay network is presented. Second, two TC algorithms based on the link EE are proposed to apply in the cluster and relay subnetworks of SDWSN, respectively. In the cluster subnetwork, the proposed distributed TC algorithm enables the link interference mitigation by employing power control and rate allocation in each cluster. In the relay subnetwork, the proposed centralized TC algorithm first utilizes a specified model to construct the original topology. During the dynamic topology maintenance, the proposed centralized TC algorithm is realized by the value-iteration learning method based on a Markov decision process (MDP) model, upon which the state-transition probability (STP) of the relay subnetwork is obtained, where the relay-network state is composed of the link, the queue, and the residual energy ratio states for all nodes in the relay subnetwork. Finally, simulation results show that both two TC algorithms can improve the corresponding subnetwork EE of time-varying SDWSN.

With the rapid development of information technology, the multisource heterogeneous data containing meaningful information have been significantly generated by various edge devices in Internet of Energy, which is one of essential foundations of many knowledge discovery tasks based on edge computing. For some complicated tasks, essential features are owned by different data sellers offering data by blockchains. With limited budgets, buying features are crucial steps in knowledge discovery tasks in electricity spot markets, especially for learning-based algorithms. However, there are lack of proper data pricing mechanisms tailored to dynamic learning processes. Besides, existing methods cannot efficiently employ edge computing servers to obtain optimal policies for selecting features according to dynamic pricing with limited budgets. To overcome such drawbacks, a data pricing mechanism is proposed in this article, which consists of static and dynamic pricing parts. Based on this mechanism, given limited budgets, a feature selection (FS) algorithm considering multiple new factors is proposed, which offers near-optimal solutions for FS at different scenarios. Numeric results show the effectiveness of the proposed algorithms.

Autonomous driving has so far received numerous attention from academia and industry. However, the inevitable occlusion is a great menace to safety and reliable driving. Existing works have primarily focused on improving the perception ability of a single autonomous vehicle (AV), but the safety problem brought by occlusions remains unanswered. In this paper, we propose a multi-tier perception task offloading framework with a collaborative computing approach, where an AV is able to achieve a comprehensive perception of the concerned region-of-interest (RoI) by leveraging collaborative computation with nearby AVs and road side units (RSUs). Besides, the collaborative computation provides offloading service for computationally intensive tasks so as to reduce processing delay. Specifically, we formulate a joint problem of perception task assignment, offloading and resource allocation, by fully considering the AV’s mobility, task dependency, and delay requirement. The collaborative offloading is modeled as a mixed-integer nonlinear programming (MINLP) problem. We design a two-layer binary intelligent firefly (TL-BIFA) algorithm to solve MINLP, with the goal of minimizing execution delay. The proposed TL-BIFA synthesizes the advantages of heuristic methods and deterministic methods. Through extensive simulations, the proposed collaborative offloading approach and the TL-BIFA show superiority in enhancing the autonomous driving system’s safety, efficiency and resource utilization.

Resistance distance is a fundamental metric to measure the similarity between two nodes in graphs which has been widely used in many real-world applications. In this paper, we study two problems on approximately computing resistance distance: (i) single-pair query which aims at calculating the resistance distance r(s, t) for a given pair of nodes (s, t); and (ii) single-source query which is to compute all the resistance distances r(s, u) for all nodes u in the graph with a given source node s. Existing algorithms for these two resistance distance query problems are often costly on large graphs. To efficiently solve these problems, we first establish several interesting connections among resistance distance, a new concept called v-absorbed random walk, random spanning forests, and a newly-developed v-absorbed push procedure. Based on such new connections, we propose three novel and efficient sampling-based algorithms as well as a deterministic algorithm for single-pair query; and we develop an online and two index-based approximation algorithms for single-source query. We show that the two index-based algorithms for single-source query take almost the same running time as the algorithms for single-pair query with the aid of a linear-size index. The striking feature of all our algorithms is that they are allowed to select an easy-to-hit node by random walks on the graph. Such an easy-to-hit landmark node v can make the v-absorbed random walk sampling, spanning tree sampling, as well as the v-absorbed push more efficient, thus significantly improving the performance of our algorithms. Extensive experiments on 5 real-life datasets show that our algorithms substantially outperform the state-of-the-art algorithms for two resistance distance query problems in terms of both running time and estimation errors.

Personalized PageRank (PPR) computation is a fundamental problem in graph analysis. The state-of-the-art algorithms for PPR computation are based on a bidirectional framework which include a deterministic forward push and a Monte Carlo sampling procedure. The Monte Carlo sampling procedure, however, often has a relatively-large variance, thus reducing the performance of the PPR computation algorithms. To overcome this issue, we develop two novel variance-reduced Monte Carlo techniques for PPR computation. Our first technique is to apply power iterations to reduce the variance of the Monte Carlo sampling procedure. We prove that conducting few power iterations can significantly reduce the variance of existing Monte Carlo estimators, only with few additional costs. Moreover, we show that such a simple and novel variance-reduced Monte Carlo technique can achieve comparable estimation accuracy and the same time complexity as the state-of-the-art bidirectional algorithms. Our second technique is a novel progressive sampling method which uses the historical information of former samples to reduce the variance of the Monte Carlo estimator. We develop several novel PPR computation algorithms by integrating both of these variance reduction techniques with two existing Monte Carlo sampling approaches, including random walk sampling and spanning forests sampling. Finally, we conduct extensive experiments on 5 real-life large graphs to evaluate our solutions. The results show that our algorithms can achieve much higher PPR estimation accuracy by using much less time, compared to the state-of-the-art bidirectional algorithms.

Achieving reliable intervehicle positioning is one of the most fundamental elements for many vehicular applications, including collision avoidance and autonomous driving. Vehicle position is generally provided by a global navigation satellite system (GNSS), which unfortunately suffers from inaccuracy to varying degrees in challenging environments, for example, GNSS outages. In this article, a reliable fusion technique, called non-Gaussian Redheffer weighted least squares ( n GRWLSs), is proposed. This new approach highlights the intervehicle positioning estimation in multi-GNSS outage environments, such as complete, partial, and free GNSS pseudorange outages. The proposed method combines, on the one hand, the benefits of the Gaussian dynamical matrix principle and the Redheffer distribution function for the sparse property in complete GNSS pseudorange outages and, on the other hand, the use of the optimal window size to regulate the data flow generated by both the inertial navigation systems (INSs) and GNSS during a partial GNSS pseudorange outage. During the free GNSS pseudorange outage, the process ignores data from the INS, and instead, GNSS pseudorange information alone will be considered to compute the intervehicle positioning information. Consequently, weighted least squares is used as an intervehicle positioning estimator. To address the pseudorange uncommon and INS measurement noises, the generalized error distribution (GED) is used to estimate the non-Gaussian densities. Finally, road-test experiments are implemented to evaluate the consistency of the proposed approach. The experimental results show that the proposed n GRWLS can accurately estimate the intervehicle positioning under various conditions (free, partial, and complete GNSS pseudorange outages).

Understanding the private car aggregation effect is conducive to a broad range of applications, from intelligent transportation management to urban planning. However, this work is challenging, especially on weekends, due to the inefficient representations of spatiotemporal features for such aggregation effect and the considerable randomness of private car mobility on weekends. In this article, we propose a deep learning framework for a spatiotemporal attention network (STANet) with a neural algorithm logic unit (NALU), the so-called STANet-NALU, to understand the dynamic aggregation effect of private cars on weekends. Specifically: 1) we design an improved kernel density estimator (KDE) by defining a log-cosh loss function to calculate the spatial distribution of the aggregation effect with guaranteed robustness and 2) we utilize the stay time of private cars as a temporal feature to represent the nonlinear temporal correlation of the aggregation effect. Next, we propose a spatiotemporal attention module that separately captures the dynamic spatial correlation and nonlinear temporal correlation of the private car aggregation effect, and then we design a gate control unit to fuse spatiotemporal features adaptively. Further, we establish the STANet-NALU structure, which provides the model with numerical extrapolation ability to generate promising prediction results of the private car aggregation effect on weekends. We conduct extensive experiments based on real-world private car trajectories data. The results reveal that the proposed STANet-NALU outperforms the well-known existing methods in terms of various metrics, including the mean absolute error (MAE), root mean square error (RMSE), Kullback-Leibler divergence (KL), and R2.

Spatio-temporal forecasting has a wide range of applications in smart city efforts, such as traffic forecasting and air quality prediction. Graph Convolutional Recurrent Neural Networks (GCRNN) are the state-of-the-art methods for this problem, which learn temporal dependencies by RNNs and exploit pairwise node proximity to model spatial dependencies. However, the spatial relations in real data are not simply pairwise but sometimes in a higher order among multiple nodes. Moreover, spatio-temporal sequences deriving from nature are often regulated by known or unknown physical laws. GCRNNs rarely take into account the underlying physics in real-world systems, which may result in degenerated performance. To address these issues, we devise a general model called Mixed-Order Relation-Aware RNN (MixRNN+) for spatio-temporal forecasting. Specifically, our MixRNN+ captures the complex mixed-order spatial relations of nodes through a newly proposed building block called Mixer, and simultaneously addressing the underlying physics by the integration of a new residual update strategy. Experimental results on three forecasting tasks in smart city applications (including traffic speed, taxi flow, and air quality prediction) demonstrate the superiority of our model against the state-of-the-art methods. We have also deployed a cloud-based system using our method as the bedrock model to show its practicality.

Modern sociology has profoundly uncovered many convincing social criteria for behavioral analysis. Unfortunately, many of them are too subjective to be measured and very challenging to be presented in online social networks (OSNs) for the large data volume and complicated environments to be explored. On the other hand, data mining techniques can better find data patterns but many of them leave behind unnatural understanding to humans. Although there are some works trying to integrate social observations for specific tasks, they are still hard to be applied to more general cases. In this paper, we propose a fundamental methodology to support the further fusion of data mining techniques and sociological behavioral criteria. Our highlights are three-fold: First, we propose an effective hypergraph awareness and a fast line graph construction framework. The hypergraph can more profoundly indicate the interactions between individuals and their environments because each edge in the hypergraph (a.k.a hyperedge) contains more than two nodes, which is perfect to describe social. A line graph treats each social environment as a super node with the underlying influence between different environments. In this way, we go beyond traditional pair-wise relations and explore richer patterns under various sociological criteria; Second, we propose a novel hypergraph-based neural network to learn social influence flowing from users to users, users to environments, environment to users, and environments to environments. The neural network can be learned via a task-free method, making our model very flexible to support various data mining tasks and sociological analysis; Third, we propose both qualitative and quantitive solutions to effectively evaluate the most common sociological criteria like social conformity, social equivalence, environmental evolving and social polarization. Our extensive experiments show that our framework can better support both data mining tasks for online user behaviors and sociological analysis.

While spatial-temporal environment monitoring has become an indispensable way to collect data for enabling smart cities and intelligent transportation applications, the cost to deploy, operate and maintain a sensor network with sensors and massive communication infrastructure is too high to bear. Compared to the infrastructure-based sensing approach, community sensing, or namely mobile crowdsensing, that leverage community members' mobile devices to collect data becomes a feasible way to scale up the spatial-temporal coverage of the sensing system. However, a community sensing system would need to aggregate sensors and location data from community members and thus would raise concerns on privacy and data security In this paper, we present a novel community sensing paradigm AFCS -Sensor and Location Data Aggregation-Free Community Sensing, which is designed to obtain the environment information (e.g., spatial-temporal distributions of air pollution, temperature, and bike-shares) in each subarea of the target area, without aggregating sensor and location data collected by community members. AFCS proposes to orchestrate with the Trusted Execution Environments (TEEs) of every community member's mobile device to cover the communication, computation and storage with spatial-temporal data. Further, AFCS proposes a novel Decentralized Spatial-Temporal Compressive Sensing framework based on Parallelized Stochastic Gradient Descent. Through learning the latent structure of the spatial-temporal data via decentralized optimization, AFCS approximates the value of the sensor data in each subarea (both covered and uncovered) for each sensing cycle using the sensor data locally stored in every member's TEE instance. Experiments based on real-world datasets and the Virtual Mobile Infrastructure (VMI) with TEE emulations demonstrate that AFCS exhibits low approximation error (i.e., less than 0:2°C in city-wide temperature sensing, 10 units of PM2.5 index in urban air pollution sensing, and 2 bikes in city-wide bike sharing prediction) and performs comparably to (sometimes better than) state-of-the-art algorithms based on the data aggregation and centralized computation

Device-to-device (D2D)-assisted decentralized learning has been proposed for mobile devices to collaboratively train artificial intelligence networks without the centralized parameter server. However, a densely connected network will cause large learning latency and energy consumption due to the limited computation and communication resources. In addition, link selection and aggregation weight have a significant impact on the learning performance. To cope with these challenges, we propose a joint computing power adjustment, wireless resource allocation, link selection, and aggregation weight adaptation mechanism to improve both communication and energy efficiencies. Specifically, the learning performances including the convergence rate, per-iteration learning latency, and per-iteration energy consumption are first analyzed. Then, an optimization problem is formulated to minimize the total learning cost, which is defined as the weighted sum of total learning latency and energy consumption. Given a network topology, the computing power and wireless resource allocation are optimized by the alternating optimization algorithm. Moreover, the optimal aggregation weight is obtained by semidefinite programming. With respect to link selection, we propose a tabu search based meta-heuristic algorithm to approximately achieve feasible solutions with a low computational complexity. Finally, extensive experiments demonstrate that the proposed link selection algorithm can significantly reduce the learning cost under the given learning accuracy requirement.

Providing quality communications under adversarial electronic attacks, e.g., broadband jamming attacks, is a challenging task. Unlike state-of-the-art approaches which treat jamming signals as destructive interference, this paper presents a novel active anti-jamming (AAJ) scheme for a jammed channel to enhance the communication quality between a transmitter node (TN) and receiver node (RN), where the TN actively exploits the jamming signal as a carrier to send messages. Specifically, the TN is equipped with a programmable-gain amplifier, which is capable of re-modulating the jamming signals for jamming modulation. Considering four typical jamming types, we derive both the bit error rates (BER) and the corresponding optimal detection thresholds of the AAJ scheme. The asymptotic performances of the AAJ scheme are discussed under the high jamming-to-noise ratio (JNR) and sampling rate cases. Our analysis shows that there exists a BER floor for sufficiently large JNR. Simulation results indicate that the proposed AAJ scheme allows the TN to communicate with the RN reliably even under extremely strong and/or broadband jamming. Additionally, we investigate the channel capacity of the proposed AAJ scheme and show that the channel capacity of the AAJ scheme outperforms that of the direct transmission when the JNR is relatively high.

The integration of millimeter-wave (mmWave) and non-orthogonal multiple access (NOMA) has attracted academic and industrial interest as a promising enabling approach due to their massive bandwidth support and greater spectral efficiency. In this paper, we consider NOMA heterogeneous network (HetNet) in mmWave communications, consisting of the macro-cell and small cell tiers connected through the wireless backhaul. Following this, we propose a user grouping algorithm to significantly simplify the user clustering process where each cluster consists of highly correlated users to suppress the inter-cluster interference. We then propose a hybrid analog/digital precoding scheme at the macro base station. The goal is to maximize the system’s total energy-efficiency (EE) in NOMA HetNet by jointly optimizing the transmission of hybrid precoding, power allocation, and bandwidth partition. Next, we formulate an optimization problem within the proposed framework under the minimum data rate and total maximum transmission power constraints. Due to the non-convexity nature of the formulated problem, it is challenging to obtain the optimal solution. As a result, we transform the original problem into a quasi-convex equivalent by analyzing its structure. Then we propose energy-efficient distributed power allocation algorithms to optimize the resource allocation of macro-cell and small cells. Finally, we exhibit the simulation results to demonstrate the proposed algorithm’s performance gains. The simulation results reveal that the proposed algorithmic approach surpasses state-of-the-art and traditional mmWave orthogonal multiple access systems in both capacity and EE.

Edge computing has been introduced as a promising technology for real-time video streaming systems. However, due to the lack of automatic incentives and the limitation of resources, traditional edge computing performs poorly in nowadays scenarios. To handle these two challenges, we propose a framework ofJointIncentive design andResourceAllocation (JIRA) for edge-based real-time video streaming systems. Technically, to ensure the trust and automatic distribution of incentives, we develop a novel smart contract based incentive mechanism and implement a prototype. Meanwhile, we propose an efficient online algorithm, i.e., JIRA, which dynamically adjusts compression ratio, offloading decision, and resource allocation to achieve performance optimization for video streaming under long-term latency and resource constraints. Specifically, JIRA is based on Lyapunov optimization, which decomposes the challenging long-term decision problem into a series of real-time optimization problems. Then we propose a multi-cut Generalized Benders Decomposition based algorithm (MGA) to tackle the non-convexity of the decomposed problem. Through rigorous theoretical analysis, we prove the performance bound of JIRA. Extensive simulations demonstrate that the proposed schemes can achieve an efficient trade-off between accuracy performance and energy consumption.

Transformers have been an efficient alternative to recurrent neural networks in many sequential learning tasks. When adapting transformers to modeling trajectories, we encounter two major issues. First, being originally designed for language modeling, transformers assume regular intervals between input tokens, which contradicts the irregularity of trajectories. Second, transformers often suffer high computational costs, especially for long trajectories. In this paper, we address these challenges by presenting a novel transformer architecture entitled TrajFormer. Our model first generates continuous point embeddings by jointly considering the input features and the information of spatio-temporal intervals, and then adopts a squeeze function to speed up the representation learning. Moreover, we introduce an auxiliary loss to ease the training of transformers using the supervision signals provided by all output tokens. Extensive experiments verify that our TrajFormer achieves a preferable speed-accuracy balance compared to existing approaches.

In this paper, we focus on visible light-based position and pose detection (VLP) for user devices in dynamical environments. Traditional model-based VLP methods usually depend on a perfect signal propagation model (SPM) with fixed parameters, and hence their performance will be seriously decreased when localization environment varies over time, e.g., due to diffuse scattering and reflections. To address this challenge, in this paper we propose a novel branch-structured residual convolutional neural network (RCNN)-based VLP method, without any requirement on perfect SPM knowledge. We observe that there are environment-invariant texture features in received visible light signal samples, which can be exploited for VLP performance enhancement. A branch-structured RCNN-based VLP scheme is devised for exploiting diverse-level stable texture features from received measurement samples, rendering a reliable VLP solution against environmental dynamics. It is verified by simulations that our branch-structured RCNN-based VLP solution outperforms existing machine learning-based VLP methods.

Data visualization is a powerful tool for understating information through visual cues. However, allowing novices to create visualization artifacts for what they want to see is not easy, just as not everyone can write SQL queries. Arguably, the most natural way to specify what to visualize is through natural language or speech, similar to our daily search on Google or Apple Siri, leaving to the system the task of reasoning about what to visualize and how.

Over the last decade, machine learning (ML) and deep learning (DL) algorithms have significantly evolved and been employed in diverse applications, such as computer vision, natural language processing, automated speech recognition, etc. Real-time safety-critical embedded and Internet of Things (IoT) systems, such as autonomous driving systems, UAVs, drones, security robots, etc., heavily rely on ML/DL-based technologies, accelerated with the improvement of hardware technologies. The cost of a deadline (required time constraint) missed by ML/DL algorithms would be catastrophic in these safety-critical systems. However, ML/DL algorithm-based applications have more concerns about accuracy than strict time requirements. Accordingly, researchers from the real-time systems (RTSs) community address the strict timing requirements of ML/DL technologies to include in RTSs. This article will rigorously explore the state-of-the-art results emphasizing the strengths and weaknesses in ML/DL-based scheduling techniques, accuracy versus execution time tradeoff policies of ML algorithms, and security and privacy of learning-based algorithms in real-time IoT systems.

Locating a device is a basic element for many Internet of Things (IoT) applications. In particular, it often demands an algorithm having low complexity to limit the energy consumption and most important, sufficient robustness without knowing the device in the near-field for point localization or in the far-field for direction of arrival (DOA) estimation. This article proposes a new localization algorithm that can achieve the two purposes, with the theoretical analysis to validate the optimal accuracy and the real data experiment to support the promising performance. The first objective is achieved by a closed-form solution and the second is accomplished by using the modified polar representation (MPR) of the source position, based on a new formulation for the localization problem. While the MPR localization method has been introduced before, it is not sufficiently robust for IoT application to handle the large equal radius (LER) scenario or the presence of sensor position errors. The proposed algorithm uses a different MPR formulation, which is able to handle the LER scenario, sensor position errors, and has low computational complexity.

With the development of the Internet of energy, more and more participants share data by different types of edge devices. However, such multi-source heterogenous data typically contain low-quality data, e.g., missing values, which may result in potential risks. Besides, resource-constrained devices incur large latency in edge computing networks. To alleviate such latency, distributed task offloading schemes are designed to share the computation burden between edge nodes and nearby servers. However, there are three main drawbacks of such schemes. First, low-quality data are not carefully evaluated by constraints under scenarios, which may result in slow convergence in distributed computation. Second, multi-source data including sensitive information are computed and shared among edge nodes without privacy protection. Third, distributed tasks on low-quality data may result in low-quality results even with an optimal offloading scheme. To address the problems above, a task offloading framework for edge computing based on consortium blockchain and distributed reinforcement learning is proposed in this paper, which can provide high-quality task offloading policies with data privacy protected. This framework consists of three key components: data quality evaluation (DQ) with multiple data quality dimensions, data repairing (DR) with a repairing algorithm based on a novel repairing consensus mechanism and distributed reinforcement learning for task arrangement (DELTA) with a distributed reinforcement learning algorithm based on a novel low-quality data distributing strategy. Numeric results are presented to illustrate the effectiveness and efficiency of the proposed task offloading framework for edge computing on low-quality data in the IoE.

In this paper, we investigate the precoder design for multiple-input multiple-output (MIMO) dual-functional radar-communication (DFRC) system with one-bit digital-to-analog converters (DACs). In order to form the dual-functional beam-pattern, we formulate the precoding problem as a weighted optimization problem with the constant modulus constraint, which aims at minimizing the average error power and guaranteeing radar waveform similarity. The problem is divided into three sub-problems corresponding to the multiple variables, i.e., the precoding factor, transmit signal matrix, and radar waveform matrix. Due to the discrete and non-convex properties of the optimization problem, we propose a multi-variable alternating minimization (MVAM) framework to achieve the near-optimal solutions. The precoding factor and radar waveform can be solved in closed-forms. For the transmit signal matrix, we devise a binary particle swarm optimization-simulated annealing (BPSO-SA) algorithm to obtain it under the MVAM framework. Extensive simulations validate the effectiveness of the proposed approach under various scenarios, including the case without perfect channel state information. The simulation results show that, compared with existing non-linear precoders, the proposed approach achieves 7dB SNR gain at the bit error rate of 10−4in the 8-antenna system, and the gain of SNR is 0.2dB in the massive MIMO system with 128 antennas.

Hyperspectral images (HSIs) have gained high spectral resolution due to recent advances in spectral imaging technologies. This incurs problems, such as an increased data scale and an increased number of bands for HSIs, which results in a complex correlation between different bands. In the applications of remote sensing and earth observation, ground objects represented by each HSI pixel are composed of physical and chemical non-Euclidean structures, and HSI classification (HIC) is becoming a more challenging task. To solve the above problems, we propose a framework based on a deep attention graph convolutional network (DAGCN). Specifically, we first integrate an attention mechanism into the spectral similarity measurement to aggregate similar spectra. Therefore, we propose a new similarity measurement method, i.e., the mixed measurement of a kernel spectral angle mapper and spectral information divergence (KSAM-SID), to aggregate similar spectra. Considering the non-Euclidean structural characteristics of HSIs, we design deep graph convolutional networks (DeepGCNs) as a feature extraction method to extract deep abstract features and explore the internal relationship between HSI data. Finally, we dynamically update the attention graph adjacency matrix to adapt to the changes in each feature graph. Experiments on three standard HSI data sets, namely, the Indian Pines, Pavia University, and Salinas data sets, demonstrate that the DAGCN outperforms the baselines in terms of various evaluation criteria. For example, on the Indian Pines data set, the overall accuracy of the proposed method achieves 98.61% when the training sample is 10%.

The explosive growth of road vehicles especially the private cars has brought unprecedented pressure to a series of problems in urban transportation systems, such as traffic congestion and environmental pollution. Private cars trajectory data and perceiving their information provide a promising solution to these problems. However, the collection of large-scale trajectory data for private cars with high accuracy and reliability is still delicate tasks in urban environments. In this paper, we propose a low-cost and user-friendly implementation method for achieving large-scale private cars trajectory data acquisition via designing lightweight GPS module and On Board Diagnostics (OBD) reader. To ensure reliable trajectory data acquisition via GPS/OBD integration, we propose an ensemble learning based Gauss Process Regression (GPR) method so as to cope with the non-linearity, non-stationarity and incremental training problems during trajectory collection. We design a classification-type loss (CTL) function and build a regression to classification (R2C) method with Learn++ for realizing ensemble learning. The proposed approach implements incremental learning when new trajectory data arrives and is able to resolve the concept drifting problem. Experiments in real-world urban environment have demonstrated the effectiveness and reliability of the proposed method, it achieves better trajectory prediction performance than the comparative methods under various road conditions in GPS-denied areas.

With the fast development of urbanization and motorization, an increasing number of people choose to buy private cars to fulfill their daily travel needs. In particular, many people from various positions of the city drive their cars to specified areas, and then they will stop and stay for a certain period of time, leading to a spatiotemporal evolution of urban area attractiveness (AA). In this paper, we aim at understanding urban AA based on analyses of private car trajectory datasets. Specifically, by extracting point-of-stop (PoS) data from the private car trajectories, we design the variational Bayesian Gaussian mixture models (VBGMM) to deduce the probability density distribution of PoSs and connect it to the variation of AA. We establish a deep learning model based on long short-term memory (LSTM) to capture the evolution of the AA. Furthermore, we integrate dropout in the LSTM method to address challenging issues such as overfitting and time-consuming training of complex neural networks in the AA prediction. We conduct experiments by using real-world private car trajectory data to evaluate the performance of the proposed method. The results validate that our proposed method outperforms existing ones in terms of various metrics. To the authors’ knowledge, our work is the first one to utilize private car trajectory data to study urban area attractiveness, thereby facilitating a new perspective regarding an understanding of human travel behavior and the evolution of urban mobility.

Ambient backscatter communication (AmBC), whose passive tags transmit information to readers through radio frequency (RF) signals in the air, has attracted much attention due to its promising prospects in green internet of things (IoTs). However, AmBC detection has faced a new challenge at the reader because the received signal is mainly a hybrid signal from direct link and backscatter link, which makes it difficult to detect symbols of the backscatter tag. To address this problem, in this paper, frequency diverse array (FDA) is utilized to transmit time-variant ambient signal, which can separate the two link signals received by the reader. We propose an adaptive dual-threshold detector by employing the time-variant characteristic of FDA RF signal, which is based on the position of the boundary backscatter tag defined. Specifically, the closed form expressions for maximum likelihood (ML) threshold and adaptive thresholds are derived, respectively. Furthermore, the computational complexities are analyzed. Finally, numerical results are presented to show that the proposed adaptive dual-threshold detector achieves better bit error rate (BER) performance than the conventional ML detector.

Non-orthogonal multiple access (NOMA) and millimeter wave (mmWave) are two key enabling technologies for the fifth-generation (5G) mobile networks and beyond. In this paper, we consider uplink communications with a hybrid beamforming structure and focus on improving the spectral efficiency (SE) and energy efficiency (EE) of mmWave multiple-input multiple-output (MIMO)-NOMA systems with enhanced user grouping and power allocation. It is noted that the optimization of the SE/EE is a challenging task due to the non-linear programming nature of the corresponding problem involving user grouping, beam selection, and power allocation. Our idea is to decompose the overall optimization problem into a mixed integer problem comprised of user grouping and beam selection only, followed by a continuous problem involving power allocation and digital beamforming design. Exploiting the directionality property of mmWave channels, we first propose a novel initial agglomerative nesting (AGNES) based user grouping algorithm by taking advantage of the channel correlations. To avoid the prohibitively high complexity of the brute-force search approach and to address the overlapping beam problem, we propose two suboptimal low-complexity user grouping and beam selection schemes, the two-stage direct AGNES (D-AGNES) scheme and the joint successive AGNES (S-AGNES) scheme. We also introduce the quadratic transform (QT) to recast the non-convex power allocation optimization problem into a convex one subject to a minimum required data rate of each user. The continuous problem is solved by iteratively optimizing the power and the digital beamforming. Extensive simulation results have shown that our proposed mmWave-NOMA design outperforms the conventional orthogonal multiple access (OMA) scenario and the state-of-art NOMA schemes.

Non-orthogonal multiple access (NOMA) and millimeter wave (mmWave) transmission are two key enabling technologies for the fifth-generation (5G) mobile networks and beyond. In this paper, we consider mmWave NOMA systems with max-min fairness constraints. On the one hand, existing beamforming designs aiming at maximizing the spectral efficiency are unsuitable for the NOMA systems with fairness considered in this paper. On the other hand, previous work on mmWave NOMA mostly depends on full knowledge of channel state information (CSI) which is extremely difficult to obtain accurately in mmWave communication systems. To address this problem, we propose a heuristic hybrid beamforming design based on a statistical CSI (SCSI) user grouping strategy. An analog beamforming scheme is first proposed to mitigate inter-cluster interference in a first stage. Then two digital beamforming designs are proposed to further suppress the interference based on SCSI. One is the widely used zero forcing approach and the other is derived from the signal-to-leakage-plus-noise ratio metric extended from orthogonal multiple access systems. The effective gains fed back from the users are used for power allocation. We introduce the quadratic transform method and bisection approach to reformulate this complex problem so as to render it solvable. Simulation results show that our proposed algorithms outperform existing algorithms in term of user fairness.

Sparse vector coding (SVC) is emerging as a feasible solution for short packet transmission, which can provide better reliability and lower latency. In this article, a uniquely decomposable constellation group (UDCG)-based SVC called UDCG-SVC is proposed. The important idea of UDCG-SVC is to map the bits into the nonzero elements of the sparse vector by using the sub-constellations in the UDCG. After random spreading, the resource blocks carry the unique decomposable superimposed constellations, which ensures that the minimum Euclidean distance (MED) of the superimposed constellation is large. This results in an improved block error rate (BLER) performance of SVC. Numerical results demonstrate the advanced BLER simulation of UDCG-SVC, and the proposed UDCG-SVC scheme is better than constellation rotation-based SVC (CR-SVC) scheme.

Nowadays, microgrids (MG) have attracted much attention, as a key technology of the Internet of Energy (IoE). A great deal of research have shown that the hierarchical microgrid is a more novel structure of IoE. Although the hierarchical microgrid model solves the problem of weak power scheduling capability across microgrids, it suffers from severe communications uncertainty, which can lead to communication delay and fluctuation. To obtain the accurate result of the renewable energy accommodation assessment capacity, a hierarchical microgrid model considering communication uncertainty is proposed in this article. The solution to solve the problem of the assessment renewable energy accommodation capacity for hierarchical MG is a hybrid control based on distribution deep reinforcement learning. The temporal difference (TD) generation adversarial network (TD-GAN) is proposed as a value-based method. Compared with the policy-based method, it can better solve the distributed problem in hybrid control with a generation adversarial network (GAN). Moreover, the challenge that the method cannot handle a continuous action space is solved by using a normalized advantage function (NAF). The method similar with the TD error method is employed to train the GAN network. Simulation results using real power grid data demonstrate the effectiveness and accuracy of the proposed method.

Hybrid precoding architecture, as a cost-effective approach for millimeter-wave (mmWave) communications, can achieve an excellent tradeoff between spectrum efficiency and hardware implementation complexity. However, the design of a robust hybrid precoding for improving the physical layer security (PLS), which is insensitive to the uncertainty of eavesdropper's channel state information (CSI), has not been well studied. This article for the first time designs a probabilistically robust hybrid precoding scheme for securing broadcast communications in Internet of Things (IoT) with eavesdropper's imperfect CSI. Specifically, considering the Gaussian CSI error model, we maximize the minimum secrecy rate of multiple IoT devices (IoDs) by jointly designing analog and digital precoders under the constraints in terms of secrecy outage probability and per IoD's information rate. The optimization problem is challenging due to the coupling of the analog and digital precoders, and the secrecy outage constraint. To handle these challenges, we first employ a conservative probability inequality to transform the secrecy outage probability constraint into a deterministic one. Then, by employing the penalty dual decomposition (PDD) method, we develop a novel iterative algorithm to convert the resultant nonconvex problem into a sequence of convex problems, which can guarantee the convergence to its Karush-Kuhn-Tucker (KKT) solution. Simulation results show that the proposed algorithm can achieve significant secrecy performance gains compared with the benchmark algorithm.

The operation and management of intelligent transportation systems (ITS), such as traffic monitoring, relies on real-time data aggregation of vehicular traffic information, including vehicular types (e.g., cars, trucks, and buses), in the critical roads and highways. While traditional approaches based on vehicular-embedded GPS sensors or camera networks would either invade drivers’ privacy or require high deployment cost, this article introduces a low-cost method, namely,SenseMag, to recognize the vehicular type using a pair of noninvasive magnetic sensors deployed on the straight road section.SenseMagfilters out noises and segments received magnetic signals by the exact time points that the vehicle arrives or departs from every sensor node. Furthermore,SenseMagadopts a hierarchical recognition model to first estimate the speed/velocity, then identify the length of the vehicle using the predicted speed, sampling cycles, and the distance between the sensor nodes. With the vehicle length identified and the temporal/spectral features extracted from the magnetic signals,SenseMagclassifies the types of vehicles accordingly. Some semiautomated learning techniques have been adopted for the design of filters, features, and the choice of hyperparameters. Extensive experiment based on real-word field deployment (on the highways in Shenzhen, China) shows thatSenseMagsignificantly outperforms the existing methods in both classification accuracy and the granularity of vehicle types (i.e., seven types bySenseMagversus four types by the existing work in comparisons). To be specific, our field experiment results validate thatSenseMagis with at least 90% vehicle type classification accuracy and less than 5% vehicle length classification error.

Sensor networks are an indispensable part of the Internet of Things (IoT), where sensors perform data acquisition and information processing tasks to obtain the parameters of interest so that IoT-based monitoring, diagnosis and other systems respond quickly to the changing conditions, instantaneous faults, etc. Distributed estimation algorithms are usually employed to estimate the parameters of interest in these IoT-based applications. However, when sensor networks have highly correlated input signals and nonstationary behavior in which the parameters of interest are time-varying, conventional distributed estimation algorithms suffer from severely degraded learning performance due to the large eigenvalue spread in the covariance matrix of the input signals and the random perturbation of the parameters of interest. To address these problems, this paper proposes two diffusion Bayesian subband adaptive filter (DBSAF) algorithms from a Bayesian learning perspective. As the highly-correlated input signal is whitened in a multiband structure and an estimate of the uncertainty in the parameters of interest is obtained by performing Bayesian inference, the proposed DBSAF algorithms are able to achieve better learning performance in comparison with the competing diffusion algorithms. The transient and steady-state mean square error performance of the proposed DBSAF algorithms are analyzed, and are verified by numerical simulations. A lower bound on the time-varying step-size is derived to maintain the optimal steady-state performance in nonstationary scenarios. A new method for the estimation of the noise variance is also proposed. Numerical simulations demonstrate the excellent learning performance of the proposed algorithms in comparison with benchmark algorithms.

Millimeter wave (mmWave) communications and cognitive radio technologies constitute key technologies of improving the spectral efficiency of communications. Hence, we conceive a hybrid secure precoder for enhancing the physical layer security of a cognitive mmWave wiretap channel, where a secondary transmitter broadcasts confidential information signals to multiple secondary users under the interference temperature constraint of the primary user (PU). The optimization problem is formulated as jointly optimizing the analog and digital precoder for maximizing the minimum secrecy rate of all the secondary users under practical constraints. In particular, our design satisfies the constraint on the maximum interference power received by multiple PUs, as well as the secondary users’ minimum quality-of-service (Qos), and the unit-modulus constraint on the analog precoder. Due to the non-convexity of the resultant objective function and owing to the coupling between the analog and digital precoder, the optimization problem formulated is nonconvex and nonlinear, hence it is very challenging to solve directly. Hence, we first transform it into a tractable form, and develop a penalty dual decomposition (PDD) based iterative algorithm to locate its Karush-Kuhn-Tucker (KKT) solution. Finally, we generalize the proposed PDD algorithm to a secure hybrid precoder design relying on practical finite-resolution phase shifters and show that the proposed PDD algorithm can be straightforwardly adapted to handle the scenario, where each PU is equipped with multiple antennas and the CSI of multiple eavesdroppers (Eves) is imperfectly known. Our simulation results validate the efficiency of the proposed iterative algorithm.

Phase Modulation on the Hypersphere (PMH) is a power efficient modulation scheme for the load-modulated multiple-input multiple-output (MIMO) transmitters with central power amplifiers (CPA). However, it is difficult to obtain the precise channel state information (CSI), and the traditional optimal maximum likelihood (ML) detection scheme incurs high complexity which increases exponentially with the number of transmitting antennas and the number of bits carried per antenna in the PMH modulation. To detect the PMH signals without knowing the prior CSI, we first propose a signal detection scheme, termed as the hypersphere clustering scheme based on the expectation maximization (EM) algorithm with maximum likelihood detection (HEM-ML). By leveraging machine learning, the proposed detection scheme can accurately obtain information of the channel from a few of the received symbols with little resource cost and achieve comparable detection results as that of the optimal ML detector. To further reduce the computational complexity in the ML detection in HEM-ML, we also propose the second signal detection scheme, termed as the hypersphere clustering scheme based on the EM algorithm with KD-tree detection (HEM-KD). The CSI obtained from the EM algorithm is used to build a spatial KD-tree receiver codebook and the signal detection problem can be transformed into a nearest neighbor search (NNS) problem. The detection complexity of HEM-KD is significantly reduced without any detection performance loss as compared to HEM-ML. Extensive simulation results verify the effectiveness of our proposed detection schemes.

In this article, a dynamic routing algorithm based on energy-efficient relay selection (RS), referred to as DRA-EERS, is proposed to adapt to the higher dynamics in time-varying software-defined wireless sensor networks (SDWSNs) for the Internet-of-Things (IoT) applications. First, the time-varying features of SDWSNs are investigated from which the state-transition probability (STP) of the node is calculated based on a Markov chain. Second, a dynamic link weight is designed for DRA-EERS by incorporating both the link reward and the link cost, where the link reward is related to the link energy efficiency (EE) and the node STP, while the link cost is affected by the locations of nodes. Moreover, one adjustable coefficient is used to balance the link reward and the link cost. Finally, the energy-efficient routing problem can be formulated as an optimization problem, and DRA-EERS is performed to find the best relay according to the energy-efficient RS criteria derived from the designed link weight. The simulation results demonstrate that the path EE obtained by DRA-EERS through an available coefficient adjustment outperforms that by Dijkstra's shortest path algorithm. Again, a tradeoff between the EE and the throughput can be achieved by adjusting the coefficient of the link weight, i.e., increasing the impact of the link reward to improve the EE, and otherwise, to improve the throughput.

Mobile-edge computing (MEC) has been witnessed as a promising solution for the vehicular task offloading. Due to the limited computing resource of individual MEC servers, it faces challenges when higher requirements are put forward for timely task processing of a large amount of computations in the emerging vehicular applications. In this article, we strive to realize the efficient vehicular task offloading via heat-aware MEC cooperation from the game theory perspective. Here, the heat indicates the vehicle density and is tightly related to the requests of vehicle users when they drive through the hot zones. Specifically, a deep learning-based prediction method is proposed, capturing the dynamic time-varying heat value of the hot zones based on the analysis of the real-world private car trajectory data. To identify the role of MEC in the cooperation, we take the time-delay constraint into consideration for the task offloading. To realize MEC grouping for task offloading in MEC cooperation, we formulate the MEC grouping as a utility maximization problem via designing a noncooperative game-theoretic strategy selection based on regret-matching. Furthermore, we derive the correlated equilibrium and prove that the fast convergence can be achieved. Extensive simulation results validate the effectiveness of the proposed vehicular task offloading approach under various system parameters, such as computation workload, time slots, and MEC servers number. The proposed method outperforms the existing methods, which is able to significantly reduce the task complete delay, and in the meantime enhance the MEC energy efficiency with end users' quality-of-experience guaranteed.

Proposed by Maddah-Ali and Niesen, a coded caching scheme has been verified to alleviate the load of networks efficiently. Recently, a new technique called placement delivery array (PDA) was proposed to characterize the coded caching scheme. In this paper, we consider a caching system in the scope of ultra dense networks (UDNs). Each base station (BS) has a finite cache and stores some contents. We propose an efficient coded content caching scheme called double coded caching to make the transmission robust to in-and-out wireless network quality. Then the dynamic caching and multicast scheduling are considered to jointly minimize the average delay and power of the content-centric wireless networks. This stochastic optimization problem can be formulated as a Markov decision process (MDP) with unknown transition probabilities and large state space. We propose a deep reinforcement learning approach to deal with the decision problem. Our algorithm uses a variational auto-encoder (VAE) neural network to approximate the state sufficiently, and uses a weighted double Q-learning scheme to reduce variance and overestimation of the Q function. Numerical results demonstrate that the proposed double coded caching scheme increases the probability of the successful transmission, and the caching and scheduling policy can effectively reduce the delay and the power consumption.

Private cars, a class of small motor vehicles usually registered by an individual for personal use, constitute the vast majority of city automobiles and hence significantly affect urban traffic. In particular, private cars tend to stop-and-wait (SAW) in specific regions during daily driving. This SAW behavior produces a spatiotemporal aggregation effect, which facilitates the formation of urban hot zones. In this paper, we investigate the SAW behavior and aggregation effect based on large-scale private car trajectory data. Specifically, motivated by the first law of geography, we leverage the kernel density estimation (KDE) method and extend it to three dimensions to capture the density distribution of the SAW data. Furthermore, according to the inherent relationship between the present SAW density and future SAW aggregation, we propose a 3D-KDE-based prediction model to characterize the dynamic spatiotemporal aggregation effect. In addition, we design a modified inertia weight particle swarm optimization (MIW-PSO) algorithm to determine the optimal weight coefficients and to avoid local optima during SAW prediction. Extensive experiments based on real-world private car SAW data validate the effectiveness of our method for discovering dynamic aggregation effects, therein outperforming the current methods in terms of the Kullback-Leibler (KL) divergence, mean absolute error (MAE), and root mean square error (RMSE). To the best of the authors’ knowledge, our work is the first to utilize private car trajectory data to study the aggregation effect in urban environments, thereby being able to provide new insight into the study of traffic management and the evolution of urban traffic.

In 5G wireless communication systems, it is crucial to support efficient multiplexing of enhanced mobile broadband (eMBB) and ultra-reliable and low latency (URLLC) services in the same carrier, and maximize system capacity for both services. In this paper, we mainly discuss the scenario of multiplexing of eMBB and URLLC in a new radio (NR) downlink carrier. A new reference signal is designed in which a flag signal is also used as a reference signal for the scenario of dynamically sharing the resource between eMBB and URLLC. Simulation results demonstrate that the proposed method achieves significantly improved performance comparing previous methods.

In this letter, we address the differential received signal strength based problem with model parameter errors. To deal with the model parameter errors, we adopt the robust weighted least squares criterion, leading to a minimax optimization problem. By assuming the model parameter errors lie in a ball, the minimax problem is transformed into a tractable reformulation via the S-Lemma. Confronted with the nonconvexity of the reformulated problem, we approximately solve it by applying the semidefinite relaxation. The proposed approach only requires the knowledge of the upper bounds of the model parameter errors, which are practically easy to acquire. Simulation results show that the proposed method is robust to the model parameter errors and outperforms the existing methods.

In this paper, we address the time-of-arrival (TOA) based localization problem in an adverse environment, where line-of-sight (LOS) signal propagation between the source and the sensor is not readily available, in which case we have to resort to non-line-of-sight (NLOS) signals. Two convex relaxation methods, i.e., the semidefinite relaxation (SDR) and the second-order cone relaxation (SOCR) methods, are proposed to mitigate the effect of NLOS errors on the localization performance. We consider two separate cases in which the information of the NLOS status is totally unknown and perfectly known, respectively. The proposed methods can be applied without knowing the distribution of NLOS errors. Moreover, we propose a NLOS error mitigation method that is robust to detection errors, which are generated in the process of detecting NLOS paths. Simulation results show that the proposed convex relaxation methods outperform some existing state-of-the-art methods.

For realizing robust target tracking with wireless sensor networks in the circumstance where the propagation parameters of the characteristic signal emitted by the target are unknown, a novel tracking algorithm under the particle filter framework is proposed. We propose a scheme to realize particle weight calculation without the prior knowledge about the propagation parameters of the target's characteristic signal. With the use of the monotonic relationship of the distance and the received signal strength, we define the signal characteristic sequence and particle distance sequence and utilize the modified sequence distance between the signal characteristic sequence and the particle distance sequence as the criterion to calculate the particle weight blindly with simple lightweight operations. Simulation results demonstrate the effectiveness of the proposed algorithm. Copyright © 2011 John Wiley & Sons, Ltd.

In this paper, we design a new distributed angle estimation method for localization in wireless sensor networks (WSNs) under multipath propagation environment. We employ a two-antenna anchor that can emit two linear chirp waves simultaneously, and propose to estimate the angle of departure (AOD) of the emitted waves at each receiving node via frequency measurement of the local received signal strength indication (RSSI) signal. An improved estimation method is further proposed where multiple parallel arrays are adopted to provide the space diversity. The proposed methods rely only on radio transceivers and do not require frequency synchronization or precise time synchronization between the transceivers. More importantly, the angle is estimated at each sensor in a completely distributed manner. The performance analysis is derived and simulations are presented to corroborate the proposed studies.

One critical challenge for accurate localization with Received Signal Strength Indicator (RSSI) is the anisotropic environment, which causes the RSS-Distance Relationship (RDR) to vary spatially. To alleviate localization error caused by RDR anisotropy, most of existing works adopt multiple RDR algorithms. However, we have found that the arbitrary RDR selection in these algorithms can lead to large localization error. Moreover, localization accuracy can be further enhanced by utilizing information provided by more Access Points (APs). To address these problems, we propose a Probabilistic Neural Network based localization algorithm in this paper. The algorithm features two steps: Global Optimization and Regional Compensation, during which all APs exchange information about the Blind Node (BN) to locate it collaboratively. Simulation result shows that the proposed algorithm can achieve a localization accuracy 35% higher than that of multiple RDR algorithms.

In this paper, we propose an orthogonal space-time block coding (STBC) for continuous phase modulation (CPM) over frequency-selective channels. The new method maintains the constant envelope and the phase continuity of the CPM waveforms by inserting three blocks of data-dependent symbols in addition to a cyclic prefix (CP). At the transmitter, the data blocks are precoded and transmitted in two antennas. And at the receiver, by applying the Laurent decomposition of CPM signals, only linear processing is needed to decode the CPM signal and the decoding complexity is much lower than traditional schemes. Simulation results corroborate that the proposed scheme achieves near-optimum error performances.

An unknown-position sensor can be localized if there are three or more anchors making time-of-arrival (TOA) measurements of a signal from it. However, the location errors can be very large due to the fact that some of the measurements are from non-line-of-sight (NLOS) paths. In this paper, a semi-definite programming (SDP) based node localization algorithm in NLOS environments is proposed for ultra-wideband (UWB) wireless sensor networks. The positions of sensors can be estimated using the distance estimates from location-aware anchors as well as other sensors. However, in the absence of line-of-sight (LOS) paths, e.g., in indoor networks, the NLOS range estimates can be significantly biased. As a result, the NLOS error can remarkably decrease the location accuracy, and it is not easy to accurately distinguish LOS from NLOS measurements. According to the information known about the prior probabilities and distributions of the NLOS errors, three different cases are introduced and the respective localization problems are addressed. Simulation results demonstrate that this algorithm achieves high location accuracy even for the case in which NLOS and LOS measurements are not identifiable.

ABSTRACT For wireless multiple‐input multiple‐output (MIMO) communications systems, both channel estimation error and spatial channel correlation should be considered when designing an effective signal detection system. In this paper, we propose a new soft‐output MMSE based Vertical Bell Laboratories Layered Space‐Time (V‐BLAST) receiver for spatially‐correlated Rician fading MIMO channels. In this novel receiver, not only the channel estimation errors and channel correlation but also the residual interference cancellation errors are taken into consideration in the computation of the MMSE filter and the log‐likelihood ratio (LLR) of each coded bit. More importantly, our proposed receiver generalizes all existing soft‐output MMSE V‐BLAST receivers, in the sense that, previously proposed soft‐output MMSE V‐BLAST receivers can be derived as the reduced forms of our receiver when the above three considered factors are partially or fully simplified. Simulation results show that the proposed soft‐output MMSE V‐BLAST receiver outperforms the existing receivers with a considerable gain in terms of bit‐error‐rate (BER) performance. Copyright © 2011 John Wiley & Sons, Ltd.

Abstract Benefitting from its ability to estimate the target state's posterior probability density function (PDF) in complex nonlinear and non‐Gaussian circumstance, particle filter (PF) is widely used to solve the target tracking problem in wireless sensor networks. However, the traditional PF algorithm based on sequential importance sampling with re‐sampling will degenerate if the latest observation appear in the tail of the prior PDF or if the observation likelihood is too peaked in comparison with the prior. In this paper, we propose an improved particle filter which makes full use of the latest observation in constructing the proposal distribution. Thequality prediction functionis proposed to measure the quality of the particles, and only the high quality particles are selected and used to generate the coarse proposal distribution. Then, acentroid shift vectoris calculated based on the coarse proposal distribution, which leads the particles move towards the optimal proposal distribution. Simulation results demonstrate the robustness of the proposed algorithm under the challenging background conditions. Copyright © 2010 John Wiley & Sons, Ltd.

In sparse mobile social networks, the moving activity of nodes always happen in a specific area, which corresponds to some specific community. How to guarantee the higher packet delivery ratio while reducing forwarding delay in such networks, is a challenging issue and has not been widely investigated yet. Recently, additional super-node was introduced to ferry packets between the isolated areas. However, most existing solutions assume the super-node is always moving according to the fixed trajectory. In this paper, we put some special mobile nodes in the networks, they are called postmen, and their responsibility is to carry packets for normal nodes which belong to specific communities. Our work focus on the optimization of the moving trajectory by considering the minimum transmission delay. We formulate the optimal issue into semi-Markov Decision Process model. The decision process includes two parts: Packets-choosing strategy and trajectory of packet-ferrying-determination strategy. By maximizing the individual reward of a postman, its optimal trajectory will be found. Furthermore, the proposed solution guarantees the packet delivery ratio and delay for isolated communities. Simulation results show that the proposed packet-ferrying solution outperforms two existing ferrying solutions in terms of packet delivery ratio.

In this paper, we develop a new subspace based multiuser carrier frequency offset (CFO) estimation scheme for multi-input multi- output (MIMO) orthogonal frequency division multiple access (OFDMA) uplink transmission. We exploit the rank reduction approach by equipping multiple antennas at the receiver, in which the CFO of each user is derived blindly using one dimension (1-D) search individually. The proposed scheme supports the generalized subcarrier assignment scheme and full loaded transmission with all subcarriers being allocated to users. Numerical results are provided to corroborate the proposed studies.

In this paper, we develop a distributed space-time coding scheme to cope with both time and frequency asynchronism in two-way relay network (TWRN). The convolutional coding is employed to handle multiple timing errors in the networks. We prove that both the cooperative and the multipath diversities can be achieved by linear receivers, e.g., linear zero-forcing (ZF) or minimum mean square error (MMSE) receivers. Moreover, the diversity can be guaranteed almost surely under frequency asynchronism. Numerical results are then provided to corroborate the proposed studies.

Sink scheduling is shown to be a promising scheme in wireless sensor networks. However, previous approaches on this topic suffer from poor performance due to lack of joint considerations. Therefore, in this paper, we aim to fill in the research blank. First, we develop a novel notation Placement Pattern (PP) to bound time-varying routes with placement of sinks. This bounding technique transforms time-related constraints into pattern-based ones and allows us to formulate this optimization in pattern domain. If there is only one sink, we develop a polynomial time algorithm to solve it optimally. If there are multiple sinks, we develop a column generation based approach to solve it efficiently. Simulations not only demonstrate the efficiency of proposed algorithms but also substantiate the importance of sink mobility for energy-constrained sensor networks.

We consider the problem of distributed particle filtering, where a set of nodes are required to collectively estimate the state of a nonlinear dynamic system from their individual measurements. An efficient distributed particle filters based on diffusion strategy is proposed. In diffusion strategy, nodes communicate with their one-hop neighbor nodes only, and the information is diffused across the network. The performance analysis for this distributed target tracking algorithm is given in the paper. Simulation results demonstrate the effectiveness of our algorithm.

In traditional delay tolerant networks (DTNs), there exists a potential assumption that the nodes are willing to help others for packet forwarding. However, in the real application scenarios, such as civilian DTNs, selfish behaviors always widely exist. Therefore, the assumption that nodes are cooperative is not realistic in all applications. Currently, most of the existing incentive mechanism focuses on individual selfish behaviors. Few research work is proposed on social selfish behavior in DTNs. In this paper, we stimulate the nodes to cooperate with others by using a virtual bank mechanism. This incentive mechanism can effectively avoid individual selfish behaviors. Meanwhile, we observe that under this individual selfish incentive mechanism, the social distribution is unfair. That means the poverty nodes would appear in the networks, and become the internal threats for the social DTNs. To avoid this, we introduce the Gini coefficient to measure the inequality of the social distribution. Furthermore, by using the taxation strategy, we avoid the internal threats caused by social selfishness. To demonstrate the selfish behavior, we introduce the forwarding protocol which is based on social relations of nodes. We verify the proposed methods using simulation evaluations.

Distributed beamforming is a form of collaborative communication, in which wireless nodes adjust the phases of their transmitted signals so that the signals are combined coherently at the destination. This enhances the received signal power and extends the communication range. In this paper, we propose a transmit beamforming method that does not require any explicit feedback information from the receiver for the phase adjustment. Instead, sensors use phase-locked loops (PLL) to synchronize their carriers to the received reference signal broadcast by the destination / receiver. Based on the fact that the change of phase over some distance depends not only on the distance but also on the frequency, the destination indirectly adjusts the sensors'phase by changing the frequency of the reference signal. The destination scans all the frequencies within a feasible range to find the optimal frequency at which the received power gets its maximum. The advantage of this method is that it does not require any direct interaction among the sensors and feedback between the sensors and the destination. Every sensor adjusts its own phase independently and simultaneously. Therefore, the proposed method has good scalability and is suitable for the distributed system. We also derive some performance analysis of the method and present simulation results that validate the analysis.

In this paper, we propose a distributed angle estimation method for wireless sensor network localization with multipath fading. The multiple antenna system is equipped at the anchor and the angle of departure (AOD) can be estimated individually at each sensor node with single antenna system. Further, we exploit the space diversity to improve the estimation performance by deploying multiple parallel arrays at the anchor. Analysis and simulations demonstrate the effectiveness of our method.

Contention-based geographic forwarding (CGF) is a state-free communication paradigm for data delivery in multihop wireless sensor networks. CGF is robust to frequent topology changes, scalable to large-scale node deployment and applicable to data-centric applications and resource constrained networks. However, CGF may experience significant performance degradation under unreliable links. In this work, we present the intelligent CGF (ICGF) to combat the channel variation. IGCF combines the advantages of both cooperative and contention-based forwarding, involving multiple neighbours of the sender into the local forwarding to improve the transmission reliability. ICGF differs from existing work in that it extends the cooperation scope intelligently, by sending one additional control message on demand. For this reason, the probability of cooperation void in ICGF is decreased and the single-hop packet progress is increased. The authors conduct extensive simulations to study the performance of the proposed ICGF compared with existing protocols. Simulation results demonstrate that ICGF improves the end-to-end data delivery delay, energy efficiency and data delivery ratio.

We consider the source localization problem using time-difference-of-arrival (TDOA) measurements in sensor networks. The maximum likelihood (ML) estimation of the source location can be cast as a nonlinear/nonconvex optimization problem, and its global solution is hardly obtained. In this paper, we resort to the Monte Carlo importance sampling (MCIS) technique to find an approximate global solution to this problem. To obtain an efficient importance function that is used in the technique, we construct a Gaussian distribution and choose its probability density function (pdf) as the importance function. In this process, an initial estimate of the source location is required. We reformulate the problem as a nonlinear robust least squares (LS) problem, and relax it as a second-order cone programming (SOCP), the solution of which is used as the initial estimate. Simulation results show that the proposed method can achieve the Cramer-Rao bound (CRB) accuracy and outperforms several existing methods.

We consider a two-way relay network consisting of multiple pairs of single-antenna users and multiple distributed single-antenna relays. The two communication peers in each pair of users transmit simultaneously to the relays in the first time slot, and the relays rebroadcast the received sum signal weighted by a complex gain, in the second time slot. For multi-user systems, the signal arriving at the users contains not only self interference from the back-propagation of user signals, but also inter-pair interferences from other pairs of users. In this paper, we use zero-forcing (ZF) to cancel the inter-user interference, assuming that channel-state information for all relay-peer connections are known at every relay, but no data exchange occurs between relays. We also derive two closed-form expressions for zero-forcing beamforming weights, corresponding to two different relay power constraints, which can be implemented in a distributed manner. The first approach uses standard ZF to null out every inter-pair interference and the second approach sets the total inter-pair interference to zero. We also derive a closed-form upper bound of the achievable sum-rate and show that both methods achieve the same multiplexing gain when the number of relays N is sufficient for perfect zero-forcing, namely 2K2+ K, where K is the number of user pairs. For the case of insufficient number of relays, we also propose two solutions for beamforming weights, i.e., based on diagonal loading and use of the pseudo-inverse, and compare their advantages and weaknesses.

For the problem of low altitude target tracking with acoustic wireless network, the signal propagation time delay effect must be considered. The target has been far away from its emitting position when the signal is received by sensors. This effect leads to synchronous sensors in the measurement space sample asynchronously in the state space and the sample frequency becomes unknown and time varying. In this paper, a batch type distribution fusion algorithm is proposed which consists of two steps. First, a linear search method is used for estimating the time-varying state transition time which is the parameter of least square solution for the initial state. This initial state is used again to optimize the state transition time until the iteration termination condition is satisfied. Second, a time register procedure and a distribution fusion technique are presented to obtain the global track. Simulation results verify the efficiency of the proposed method.

Packet level coding schemes are used to improve the transmission reliability and efficiency in data broadcasting applications, especially for wireless networks. However, existing coding schemes have large redundancy for packets are coded by a random way or some certain probability distributions. Although the Maximum Distance Separable (MDS) codes are designed without redundancy, they are not effective when the packet erasure probability is high. To reduce the coding redundancy, lots of work use the receiver side information to adjust the codes construction on-the-fly, which is more beneficial. However, these work have disadvantages:1) the coding redundancy is also large; 2) the feedback schemes have great affect on performance. In this paper, we design a robust feedback scheme and a novel "Rateless Online MDS Code"(ROME) to eliminate the coding redundancy. Our contributions include that our codes are throughput optimal codes without redundancy, and the analysis on theoretical finite field size bound to achieve throughput optimal codes. We also design the finite field construction to speed up the encoding and decoding process. Finally, we compare the performance between ROME and other existing codes, like RLC, LT, RT oblivious and SLT codes, and their performance when feedbacks are erased.

Due to hardware and computational constraints, wireless sensor networks (WSNs) normally do not take measurements of time-of-arrival or time-difference-of-arrival for range-based localization. Instead, WSNs in some applications use range-free localization for simple but less accurate determination of sensor positions. A well-known algorithm for this purpose is the centroid algorithm. This paper presents a range-free localization technique based on the radical line of intersecting circles. This technique provides greater accuracy than the centroid algorithm, at the expense of a slight increase in computational load. Simulation results show that for the scenarios studied, the radical line method can give an approximately 2 to 30% increase in accuracy over the centroid algorithm, depending on whether or not the anchors have identical ranges, and on the value of DOI.

Recently, sink deployment, in the form of deploying the sink among different sites so as to leverage traffic burden, is shown to be a promising scheme to save energy and prolong network lifetime in wireless sensor networks. For this paradigm, the choice of sink sites plays a critical role in the overall system performance. In this paper, we address the optimal deployment problem for the sink in wireless sensor networks, where routing issues are naturally involved. The major contribution of this paper is the development of an efficient grid-based algorithm to solve this problem. By dividing the continuous search space into a limited number of so-called communication intersections, computational complexity has been significantly reduced. A formal proof of optimality for this algorithm is given and several interesting properties have been revealed by theoretic analysis as well as experimental results.

For wireless ad hoc or sensor networks, non-flooding, guaranteed delivery routing protocols are preferred because of limited energy. In this paper we introduce TCGR, a tree cover based geographic routing protocol for wireless networks. We assign to each node a set of short labels such that nodes are embedded in a metric space induced by one or multiple trees. Based on the embedding, we use only greedy routing to deliver packets, i.e., packets are always forwarded to the neighbor closest to the destination. Unlike many previous geographic routing protocols, we guarantee a full success ratio in finding a route to the destination, if such a route exists in the network. Moreover, each node only needs to maintain a small amount of information, which is almost surely bounded by O((logn)^2) bits, and the label size and packet header size are also bounded by O((logn)^2) bits. Simulations show TCGR can achieve remarkable performance in both path stretch and node load.

Cooperative forwarding has been considered as an effective strategy for improving the geographic routing performance in wireless sensor networks (WSNs). However, we observe that existing works do not fully utilize the forwarding opportunities provided by available neighboring nodes. In this paper, we redesign the cooperative forwarding process and present a novel cooperative contention-based forwarding (CCBF) protocol for WSNs. CCBF extends the scope of cooperation and attains the full potential of cooperative forwarding at the expense of sending one additional control message on demand. We conduct extensive simulations to study the performance of the proposed CCBF compared with existing protocols. Simulation results demonstrate that CCBF decreases the end-to-end hop counts, hence, further improves the end-to-end energy efficiency and latency. Remarkably, it provides up to 50% improvement for the packet loss ratio when the retransmission mechanism is not adopted.

Wireless sensor networks usually operate under light traffic loads. However, when an event is detected, a large volume of data may be generated and delivered to the sink. The demand for simultaneous data transmission may cause severe channel collision and thus decrease communication throughput in contention-based medium access control (MAC) protocols. In this paper, we introduce a novel receiver-centric data transmission paradigm, which takes advantage of the tree structure that is naturally formed in data collection of a sensor network to assist scheduling of channel access. On the tree structure, a receiver is able to coordinate its multiple senders' channel access so as to reduce channel contention and consequently improve communication throughput. The protocol seamlessly integrates scheduling with contention-based medium access control. In addition, to ensure reliable data transmission, we propose a sequence-based lost packet recovery scheme in a hop-by-hop recovery pattern, which could further improve communication throughput by reducing control overhead. We present the performance of our receiver-centric MAC protocol through measurements of an implementation in TinyOS on TelosB motes and extend the evaluation through ns-2 simulations. Compared with B-MAC and RI-MAC, we show the benefits of improving throughput and fairness through receiver-centric scheduling under heavy traffic loads.

For multiple-input multiple-output (MIMO) wireless communication system with minimum mean square error (MMSE) detection, power allocation between pilot and data symbols is investigated under MMSE channel estimation in this paper. We first propose a novel soft-output MMSE MIMO detector by taking the channel estimation error into consideration. Then, by random matrix theorem, we propose an effective power allocation scheme between pilot and data symbols to maximize the average lower bound of minimum post-processing signal-to-interference and noise ratio (SINR) for the MIMO system, which has equal number of transmitter and receiver antennas. Simulation results validate our power allocation scheme.

This paper deals with the partial target coverage (PTC) problem in wireless sensor networks with the objective of optimizing network lifetime. We first build a linear programming formulation, which takes total time a sensor spends on covering some targets into consideration, in order to obtain a lifetime upper bound. Then, based on the information of this formulation, we develop a sensor assignment algorithm to seek an optimal time table meeting the lifetime upper bound. A formal proof of optimality is given. We compare the proposed algorithm with a state-of-the-art algorithm: column generation approach and show that the proposed algorithm significantly outperforms in terms of computational time. Experiments have been conducted to study the effect of network parameters on network lifetime and interesting insights have been offered.

Key distribution plays an important role in wireless sensor networks (WSNs), for information must be kept secure, even in an environment with limited storage ability and low data processing speed. However, it is a challenge to work in such a harsh requirement environment, the traditional key management scheme such as key distribution center (KDC) and public key cannot be used. Key pre-distribution is a good way to address this problem in WSNs, and while several schemes have been presented before, they cannot perform well in a large scale network. In this paper, we first use node's deployment knowledge to propose a hexagon scheme, and then combine it with the bivariate-polynomial to realize a new key agreement scheme. The simulation results show that our new scheme can be used in a large scale network, is attack resistant with low memory overhead, and achieves good network connectivity. We also introduce the measure how to effectively use key pre-distribution skills in the routing protocols.

In this paper, a cooperative localization algorithm is proposed that considers the existence of obstacles in mobility-assisted wireless sensor networks (WSNs). An optimal movement scheduling method with mobile elements (MEs) is proposed to address limitations of static WSNs in node localization. In this scheme, a mobile anchor node cooperates with static sensor nodes and moves actively to refine location performance. It takes advantage of cooperation between MEs and static sensors while, at the same time, taking into account the relay node availability to make the best use of beacon signals. For achieving high localization accuracy and coverage, a novel convex position estimation algorithm is proposed, which can effectively solve the problem when infeasible points occur because of the effects of radio irregularity and obstacles. This method is the only rangefree based convex method to solve the localization problem when the feasible set of localization inequalities is empty. Simulation results demonstrate the effectiveness of this algorithm.

Underwater sensor networks (UWSNs) have been proposed for many location-dependent applications such as oceanographic data collection, pollution monitoring, mine reconnaissance, etc. Accurate node localization plays an important role in realizing the potential gains of these applications. Although many localization algorithms have been proposed for terrestrial sensor networks in recent years, it is not feasible to directly use these algorithms in UWSNs since UWSNs lack a fast and reliable communication channel. Further, due to their slow convergence speeds and high communication overhead, distributed localization algorithms designed for small-scale UWSNs are not practical for large-scale underwater sensor systems. To achieve accurate and energy efficient node localization in large-scale UWSNs, an asymmetrical round trip based localization (ARTL) algorithm is proposed in this paper. This algorithm has low computational complexity and excellent scalability. Without time synchronization, this algorithm can achieve highly accurate ranging in large-scale UWSNs. Simulation results demonstrate the effectiveness of our design in terms of both localization accuracy and energy consumption.

In this paper, we consider the range-based sensor network localization with inaccurate anchor position information. First, a novel optimization algorithm named sequential greedy optimization (SGO) algorithm is proposed, and then two distributed localization algorithms are obtained: the first is obtained by applying the SGO algorithm to a convex formulation of the localization problem, named CSGLA; while the second is obtained by applying the SGO algorithm to a nonconvex formulation of the localization problem, named NCSGLA. The CSGLA must converge globally while the NCSGLA may converge locally. Both algorithms are partially asynchronous and can be implemented in a distributed fashion in networks. We demonstrate the localization performance via simulations. Simulation results show that, 1) the CSGLA algorithm works faster than the synchronous algorithm with the same localization accuracy; 2) with a reasonably good initialization, the NCSGLA can work much better than the CSGLA.

In this paper, a cooperative localization algorithm is proposed that considers the existence of obstacles in mobility-assisted wireless sensor networks (WSNs). In this scheme, a mobile anchor (MA) node cooperates with static sensor nodes and moves actively to refine location performance. The localization accuracy of the proposed algorithm can be improved further by changing the transmission range of mobile anchor node. The algorithm takes advantage of cooperation between MAs and static sensors while, at the same time, taking into account the relay node availability to make the best use of beacon signals. For achieving high localization accuracy and coverage, a novel convex position estimation algorithm is proposed, which can effectively solve the localization problem when infeasible points occur because of the effects of radio irregularity and obstacles. This method is the only range-free based convex method to solve the localization problem when the feasible set of localization inequalities is empty. Simulation results demonstrate the effectiveness of this algorithm.

Limitation on available resources is a major challenge in wireless sensor networks. Due to high rates of unexpected node/link failures, robust data delivery through multiple hops also becomes a critical issue. In this article we present a state-free gradient-based forwarding (SGF) protocol to address these challenges. Nodes running SGF do not maintain states of neighbors or network topology and thus can scale to very large networks. Without using routing tables, SGF builds a cost field called gradient that provides each node the direction to forward data. The maintenance of gradient is purely driven by data transmissions and hence incurs little overhead. To adapt to transient channel variations and topology changes, the forwarder of a routing node is selected opportunistically among multiple candidate nodes through a distributed contention process. Simulation results show that SGF achieves significant energy savings and outperforms several existing data forwarding protocols in terms of packet delivery ratio and end-to-end delay.

In this paper, we propose a range-free cooperative localization algorithm for mobile sensor networks by combining hop distance measurements and particle filtering. In the hop distance measurement step, a differential error correction scheme is devised to reduce the positioning error accumulated over multiple hops. A backoff-based broadcast mechanism is also introduced in our localization algorithm. It efficiently suppresses redundant broadcasts and reduces message overhead. The proposed localization method has fast converges with small location estimation error. We verify your algorithm in various scenarios and compare it with conventional localization methods. Simulation results show that our proposal is superior to the state-of-the-art localization algorithms for mobile sensor networks.

Wireless sensor networks (WSNs) have been proposed for a multitude of location-dependent applications. To stamp the collected data and facilitate communication protocols, it is necessary to identify the location of each sensor. In this paper, we discuss the performance of a novel received signal strength indicator (RSSI) positioning scheme, which uses a generalized geometrical location algorithm to achieve an accurate estimation based on mean received signal strength measurements. In order to improve the network performance and address limitations of static WSNs position estimation, mobile sensors are utilized effectively and an attractive movement strategy with mobile elements is designed. The effectiveness of our approach is validated and compared with the traditional RSSI method by extensive simulations.