Federated recommendation (FR) facilitates collaborative training by aggregating local models from massive devices, enabling client-specific personalization while ensuring privacy. However, we empirically and theoretically demonstrate that server-side aggregation can undermine client-side personalization, leading to suboptimal performance, i.e., the aggregation bottleneck. This issue stems from the inherent heterogeneity across numerous clients in FR, which drives the global model to deviate from local optima. To this end, we propose FedEM, which elastically merges the global and local models to compensate for impaired personalization. Unlike existing personalized federated recommendation (pFR) methods, FedEM (1) investigates the aggregation bottleneck in FR through theoretical insights, rather than relying on heuristic analysis; (2) leverages off-the-shelf local models rather than designing additional mechanisms to boost personalization. Extensive experiments demonstrate that our method preserves client personalization during collaborative training, outperforming state-of-the-art baselines.

Federated recommendations (FRs), facilitating multiple local clients to collectively learn a global model without disclosing user private data, have emerged as a prevalent on-device service. In conventional FRs, a dominant paradigm is to utilize discrete identities to represent clients and items, which are then mapped to domain-specific embeddings to participate in model training. Despite considerable performance, we reveal three inherent limitations that can not be ignored in federated settings, i.e., non-transferability across domains, ineffectiveness in cold-start settings, and potential privacy violations during federated training. To this end, we propose a transferable federated recommendation model, TransFR, which delicately incorporates the general capabilities empowered by pre-trained models and the personalized abilities by fine-tuning local private data. Specifically, it first learns domain-agnostic representations of items by exploiting pre-trained models with public textual corpora. To tailor for FR tasks, we further introduce efficient federated adapter-tuning and post-adaptation personalization, which facilitate personalized adapters for each client by fitting local private data. We theoretically prove the advantages of incorporating adapter tuning in FRs regarding both effectiveness and privacy. Through extensive experiments, we show that our TransFR surpasses state-of-the-art FRs on transferability.

Task-oriented sampling aims to predict the importance of points of a point cloud to better serve downstream tasks, which has attracted increasing attention in the fields of computer vision and visualization in recent years. However, existing methods cannot sufficiently leverage both global saliency and local saliency cues, resulting in suboptimal performance that requires further improvement. To tackle this challenge, we propose a novel 3D point cloud sampling method inspired by the human visual perception mechanism in this study, which can effectively extract important point cloud subsets from critical regions to better adapt to downstream tasks, thereby maintaining superior sampling performance. The proposed Visual Perception-inspired 3D Point Cloud Sampling (VPI-3DPS) method simulates the human visual system’s dynamic attention-shifting strategy by combining coarse-grained attention-driven sampling with fine-grained detail preservation. This allows our approach to adaptively capture both global context and local details within point cloud data, safeguarding downstream task performance. By leveraging Gated Recurrent Units (GRUs) for long-term dependency modeling and integrating Graph Convolutional Networks (GCNs) to capture local structures, VPI-3DPS obtains an integrated representation of regional correlation and detail awareness. Extensive experiments show that VPI-3DPS outperforms existing methods. Compared to the best-performing approaches, it achieves an average increase of 1.29% in classification accuracy, an average reduction of 13.20% in registration MRE, and an average decrease of 4.29% in Chamfer Distance for reconstruction.

In recent years, the rapid advancement of multi-modal large language models has propelled the development of video-based conversation models. Due to their exceptional video understanding capabilities, there is often an expectation that these models can handle all video-related tasks, including action recognition. However, because action recognition datasets typically lack semantic information, limiting the performance of dialogue models. Additionally, as these dialogue models are designed for video understanding, they frequently overlook critical information required for action recognition—continuous motion—in their model architecture and training dataset configurations. To address these challenges, we first propose a novel two-step mapping framework based on large language models, termed “Vision-Semantics-Label” mapping, to better adapt video-based large language models for action recognition. In the first step, we proposed a visual-skeletal collaborative learning large language model (VS-LLM), which utilizes human keypoints to compensate for the missing motion details without increasing the input token length of the large language model. In the second step, we designed two mapping methods: verb noun match (VN-Match) and all text match (ALL-Match), which can effectively extract relevant action descriptions from the text. Finally, we construct semantic action recognition datasets to ensure that the training data inherently contains action details, enabling the model to better achieve action recognition. We evaluate our approach on five benchmark datasets, demonstrating the state-of-the-art performance of large language models in action recognition. The source code and dataset are publicly available at https://github.com/xiaoyu92568/VS-LLM.

Vision-based 3-D object detection is a core task for autonomous driving and intelligent transport systems. However, the idealized geometric assumptions and single-view modeling methods relied upon by existing methods are susceptible to geometric uncertainties, caused by external parameter perturbations and depth estimation noise in practical multi-scenario deployments. Building a unified and robust perception model suitable for multi-scenarios of both ego-vehicle and roadside remains challenging. To address these challenges, we propose an Uncertainty-Aware Geometric Modeling (UAGM) method that explicitly handles geometric uncertainty to achieve robust perception across multiple scenarios. We use a dual-branch architecture to establish a robust geometric foundation: the height branch introduces dynamic virtual coordinate calibration to compensate for camera extrinsic parameter perturbations in real time, while explicitly modeling height prediction uncertainty through Multi-Hypothesis Projection (MHP), thereby constructing a robust global geometric representation. Meanwhile, the depth branch integrates a Probabilistic Depth Smoothing (PDS) module that employs Conditional Random Fields (CRF) to model spatial consistency constraints, effectively mitigating geometric discontinuities arising from pixel-level predictions. To facilitate better information fusion and interaction, we first propose a Temporal Pyramid Fusion (TPF) module to effectively capture multi-scale spatio-temporal dynamics to reduce the uncertainty in single-frame estimation, instead of error-prone dynamic ego-motion compensation. Subsequently, our Hierarchical Refinement Decoder (HRD) refines BEV proposal localization by fusing image features with depth embeddings to compensate for spatial distortions caused by forward projection. Experimental results demonstrate that UAGM not only achieves state-of-the-art detection performance on both the nuScenes and DAIR-V2X benchmarks, but more importantly, it successfully demonstrates the strong generalization capability of a single model across different viewpoints and deployment conditions.

This paper proposes a graph neural network (GNN)-enabled coordinated beamforming design, termed HSTGNN, for high speed railway (HSR) communication systems, where a high-speed train (HST) and low-speed users (LUs) coexist in multi-cell scenarios. Two transmission schemes with the goal of maximizing quality-of-service (QoS)-constrained sum rate and rate of the HST are formulated and then reformulated using a hybrid maximum ratio transmission and zero-forcing strategy. The HSR communication system is abstracted into a heterogeneous graph, and HSTGNN consists of complex heterogeneous graph attention layers and fully-connected layers. To meet QoS requirements and power budget constraints, we employ constraint-based penalty terms and a numerical scaling operation. HSTGNN is trained via unsupervised learning to solve the two schemes in a unified framework. Numerical results demonstrate that HSTGNN achieves millisecond-level inference speed with an average optimality gap of only 4% relative to traditional optimization algorithm across various scenarios. Moreover, HSTGNN exhibits strong scalability to unseen configurations of both cells and LUs.

Federated recommendation aims to collect global knowledge by aggregating local models from massive devices, to provide recommendations while ensuring privacy. Current methods mainly leverage aggregation functions invented by federated vision community to aggregate parameters from similar clients, e.g., clustering aggregation. Despite considerable performance, we argue that it is suboptimal to apply them to federated recommendation directly. This is mainly reflected in the disparate model structures. Different from structured parameters like convolutional neural networks in federated vision, federated recommender models usually distinguish itself by employing one-to-one item embedding table. Such a discrepancy induces the challenging embedding skew issue, which continually updates the trained embeddings but ignores the non-trained ones during aggregation, thus failing to predict future items accurately. To this end, we propose a personalized Federated recommendation model with Composite Aggregation (FedCA), which not only aggregates similar clients to enhance trained embeddings but also aggregates complementary clients to update non-trained embeddings. Besides, we formulate the overall learning process into a unified optimization algorithm to jointly learn the similarity and complementarity. Extensive experiments on several real-world datasets substantiate the effectiveness of our proposed model. Our code is available at https://github.com/hongleizhang/FedCA .

Federated machine unlearning service is an emerging Machine-Learning-as-a-Service (MLaaS) paradigm which supports the request of removing or forgetting the influence of a specific group of data from federated learning services. While current federated unlearning methods work well in removing instances on individual clients, they encounter challenges in addressing multiple forms of unlearning requirements, such as heterogeneous learning models and diverse unlearning contents. In this paper, we propose novel federated few-shot learning and unlearning models inspired by distillation learning. Firstly, we propose a federated metric learning model where only prototypes are transmitted between the server and clients with limited samples. The training data with different input dimensions on local clients are transformed into abstract prototypes with the same length, enabling collaborative training of heterogeneous models among clients. Then, we propose an efficient federated metric unlearning method, where the temporarily stored prototypes are utilized as teacher knowledge to guide and accelerate the retraining process for each unlearning scenario. The complexity of the federated metric unlearning method is analyzed to show the computation time and communication efficiency. Experimental results demonstrate that our approach outperforms baseline methods in terms of accuracy while effectively removing various requested unlearning contents.

The use of neural differential equation models in machine learning applications has gained significant traction in recent years. In particular, fractional differential equations (FDEs) have emerged as a powerful tool for capturing complex dynamics in various domains. While existing models have primarily focused on constant-order fractional derivatives, variable-order fractional operators offer a more flexible and expressive framework for modeling complex memory patterns. In this work, we introduce the Neural Variable-Order Fractional Differential Equation network (NvoFDE), a novel neural network framework that integrates variable-order fractional derivatives with learnable neural networks. Our framework allows for the modeling of adaptive derivative orders dependent on hidden features, capturing more complex feature-updating dynamics and providing enhanced flexibility. We conduct extensive experiments across multiple graph datasets to validate the effectiveness of our approach. Our results demonstrate that NvoFDE outperforms traditional constant-order fractional and integer models across a range of tasks, showcasing its superior adaptability and performance.

Current collaborative perception methods often rely on fully annotated datasets, which can be expensive to obtain in practical situations. To reduce annotation costs, some works adopt sparsely supervised learning techniques and generate pseudo labels for the missing instances. However, these methods fail to achieve an optimal confidence threshold that harmonizes the quality and quantity of pseudo labels. To address this issue, we propose an end-to-end Collaborative perception Dual Teacher-Student framework (CoDTS), which employs adaptive complementary learning to produce both high-quality and high-quantity pseudo labels. Specifically, the Main Foreground Mining (MFM) module generates high-quality pseudo labels based on the prediction of the static teacher. Subsequently, the Supplement Foreground Mining (SFM) module ensures a balance between the quality and quantity of pseudo labels by adaptively identifying missing instances based on the prediction of the dynamic teacher. Additionally, the Neighbor Anchor Sampling (NAS) module is incorporated to enhance the representation of pseudo labels. To promote the adaptive complementary learning, we implement a staged training strategy that trains the student and dynamic teacher in a mutually beneficial manner. Extensive experiments demonstrate that the CoDTS effectively ensures an optimal balance of pseudo labels in both quality and quantity, establishing a new state-of-the-art in sparsely supervised collaborative perception.

Recent graph-based multi-view clustering (GMVC) methods typically encode view features into high-dimensional spaces and construct graphs based on distance similarity. However, the high dimensionality of the embeddings often leads to the hubness problem, where a few points repeatedly appear in the nearest neighbor lists of other points. We show that this negatively impacts the extracted graph structures and message passing, thus degrading clustering performance. To the best of our knowledge, we are the first to highlight the detrimental effect of hubness in GMVC methods and introduce the hubREP (hub-aware Representation Embedding and Pairing) framework. Specifically, we propose a simple yet effective encoder that reduces hubness while preserving neighborhood topology within each view. Additionally, we propose a hub-aware pairing module to maintain structure consistency across views, efficiently enhancing the view-specific representations. The proposed hubREP is lightweight compared to the conventional autoencoders used in state-of-the-art GMVC methods and can be integrated into existing GMVC methods that mostly focus on novel fusion mechanisms, further boosting their performance. Comprehensive experiments performed on eight benchmarks confirm the superiority of our method. The code is available at https://github.com/zmxu196/hubREP.

Zero-Shot Learning focuses on recognizing images from unseen classes with the model trained only on seen classes and auxiliary information. Auxiliary information represents the semantic concepts of classes and is crucial for unseen class generalization. Existing works have tried human-annotated attributes, word embedding, or texts as auxiliary information. However, these auxiliary information is semantically insufficient and visually misaligned, constraining model performance. In this work, we propose the Coarse-to-Fine-to-Fusion prototype generation network (CFF). To obtain vision-oriented text corpora, we design the Coarse-to-Fine Text Generation (CFTG) paradigm, utilizing large language models to generate coarse- and fine-grained texts. For text embedding and fusion, we propose the Semantic Prototype Generation (SPG) module, fusing learnable prompts with coarse- and fine-grained embedding, enabling fine-grained and fusion prototype generation. Moreover, we propose the Visual-Semantic Alignment (VSA) loss to narrow the domain gap. Extensive experiments on AWA2, SUN, and CUB datasets demonstrate the effectiveness of our method.

Recent advancements in 3D Large Language Models (LLMs) have revealed significant potential in enhancing the understanding of 3D scenes. However, previous methods have struggled with extracting and utilizing fine-grained information of 3D objects for the coarsness of point clouds, resulting in limitations in understanding object-of-interested (OoI) within the scene. To address this issue, we introduce the object-centric 2D-3D interaction module for enhancing the ability of LLMs for 3D understanding tasks, which consists of the fine-grained 2D representation perception and the object-centric 3D scene representation perception. Specifically, the 2D representation associated with 3D objects is captured based on cross-modal semantic consistency without any spatial projector. Experimental results show that our model significantly outperforms existing methods on benchmarks including ScanRefer and ScanQA.

Machine learning and artificial intelligence technologies have become integral to various industries, driven by the availability of large-scale data. However, the use of sensitive industrial and personal data introduces significant privacy risks. Generative Adversarial Networks (GANs) are employed to generate synthetic data, thereby rendering them a feasible privacy-preserving technique. Despite their potential, existing private GANs face two major limitations: they are restricted to single-modal data generation or fail to ensure strict privacy guarantees. To tackle these issues, we propose Differential Privacy GAN of Image and Table—DPGAN-IT, which is capable of generating both image and tabular data simultaneously while enforcing robust privacy protection. Extensive experiments validate high utility of multi-modal synthetic data and demonstrate an effective balance between privacy and data utility.

Deploying convolutional neural networks (CNNs) on resource-constrained Internet of Things (IoT) devices facilitates convenient intelligent services, which has attracted extensive attention. However, the dynamic resource availability poses significant challenges for resource-intensive CNNs. Recent advancements in efficient CNNs have two limitations: i) elevated resource consumption due to extensive datasets and post-training calibration; ii) lacking flexibility facing dynamic resources due to static model structure. To this end, we propose a RESourCe-Aware oncE-for-alL (ReScale) framework for efficient CNN deployment on IoT devices. Specifically, we propose a hierarchical filter generation mechanism to generate different amounts of filters dynamically, mapping a few learnable filters to abundant filters for discriminative feature extraction. Besides, we set a coefficient$\gamma$to modulate the filter generation according to the available resources of IoT devices. With this framework, we only need to train CNN models once to handle different resource availability of IoT devices. Experimental results show that our proposed ReScale framework can generate more efficient models with lower resource consumption while maintaining high accuracy. Through the coefficient$\gamma$, our method enables continuous model generation, ensuring robust adaptation to dynamic resource constraints.

Pair-then-Aggregate (PtA) introduces a programming paradigm and an automated parallel execution engine for large-scale secure multi-party (MPC) computations, drawing inspiration from the widely-used yet not explicitly defined Table-Generation-and-Look-up (TGL) pattern in privacy-preserving algorithm design. PtA offers an easy-to-use API and a versatile execution engine that harnesses various levels of parallelism and adapts to different MPC deployments, algorithms, and input sizes. Evaluations on a real-world MPC platform demonstrate significant enhancements in scalability, adaptability, and ease of programming. PtA can process one billion input elements with 3-23 lines of C++ code in 5-74 seconds. It outperforms state-of-the-art implementations in 91.4 % of 35 test cases, achieving up to a$12.4 \times$speedup with much less coding effort.11Our code is provided in https://github.com/Fannxy/Pair-then-Aggregate

Multimodal learning integrates diverse modalities to enhance robustness, yet real-world scenarios suffer from heterogeneous imbalance phenomena (noise interference, modality partial missing, intermodal information disparities), degrading performance through biased feature representations. Existing methods fail to adaptively modulate models under dynamic imbalance conditions. We propose GMML, a framework dynamically balancing multimodal gradients to counteract imbalance-induced biases: i) An imbalance-aware gradient modulation adaptively identifies contributions with smooth weight transitions to balance conflicting gradients; ii) A parameter constraint method enforces ℓ2-norm constraints on encoders, suppressing parameter oscillations and blocking noisy updates under modality missing/noise. Theoretically, GMML achieves a larger certified radius upper bound for complex imbalances, with convergence radius analysis providing theoretical guarantees. Experiments demonstrate superior robustness against three imbalance types, outperforming state-of-the-art by 3.3% and 2.3% in accuracy on KS and UCF-101 benchmarks. series Code: https://github.com/zhangzikai-security-ML/GMML.

Given the scarcity of real data and the time-intensive nature of labeling, current multi-agent perception models often rely on simulated sensor data for training and validation. However, perception performance deteriorates significantly due to domain gap between simulated and real data. Existing adaptation methods focus on domain-generalized feature extraction while neglecting multi-agent shift uncertainty and relational semantic loss. To address this issue, we propose a Selective Shift Domain Adaptation method in multi-agent collaborative perception, called SSDA. SSDA incorporates two essential components: the frequency-decoupled feature shift adjustment (FSA) and the entropy-driven staged adaptive alignment (SAA). To mitigate the relational semantic loss, the FSA is proposed to simplify the representation of correlation features and remove redundant information from the source domain, thereby mitigating interference for domain adversarial scenarios. To tackle the shift uncertainty, the SAA is designed to achieve adaptive alignment from global to local guided by information entropy, which dynamically adjusts weights for samples according to their level of uncertainty. The results demonstrate that the SSDA is significantly superior to the SOTA, achieving up to 7.35% improvements on [email protected].

Existing few-shot action recognition (FSAR) studies predominantly follow a metric learning framework, where prototypes are generated directly from features extracted by an encoder, and classification is performed via distance-based matching. However, due to the limited number of available samples, significant variations exist between different video features of the same class. As a result, the same query video may yield different classification results when matched against different sets of support videos. To address this issue, we propose a novel Hierarchical Meta-Prototypes Network (HMP-Net). The key innovation of our approach lies in the introduction of a category-agnostic and feature-agnostic meta-prototype module, which guides video feature mapping into a more suitable feature space. To optimize this meta-prototype, we design an alternating meta-prototype training strategy, where the model first learns to transform features under a fixed meta-prototype, and then the meta-prototype is refined to better guide feature mapping. Additionally, to adapt image-based metric learning models to video-based FSAR tasks, we introduce a series of lightweight adaptation modules. Specifically, we integrate an adapter into the encoder to improve video frame feature extraction, design a hierarchical prototype generation mechanism to enhance overall video understanding, and incorporate a task-specific perception module to extract unique features for each task. These adaptations make our model better suited for FSAR, significantly improving performance. We evaluate HMP-Net on five challenging benchmarks, and experimental results demonstrate that our model achieves new state-of-the-art performance on HMDB51, UCF101, Kinetics, and SthSthV2-Small. Extensive empirical evaluations further highlight the effectiveness and robustness of HMP-Net.

Continual Generalized Category Discovery (C-GCD) aims to incrementally identify both known and novel classes from unlabeled data streams while preserving previously acquired knowledge. However, current approaches face a critical limitation we term unstructured knowledge interference, a critical issue that arises when unconstrained parameter updates entangle discriminative representations across classes, severely contaminating the feature space and introducing significant transfer and bias risks. To address these challenges, we propose the Tree of Prompts (ToP), a novel hierarchical prompting framework that facilitates structured knowledge adaptation through multi-granular parameter regulation. ToP hierarchically integrates three synergistic components: (1) Stage-level prompts preserve historical knowledge by isolating task-specific parameters, thereby mitigating conflicts between incremental tasks; (2) Centroid-level prompts disentangle category semantics through learnable prototype calibration, sharpening decision boundaries in the feature space; and (3) Context-level prompts dynamically capture discriminative local features to suppress contamination from superficial similarities. Experimental results demonstrate that ToP markedly outperforms existing methods and provides a comprehensive and efficient solution for C-GCD.

The integration of differential equations with neural networks has created powerful tools for modeling complex dynamics effectively across diverse machine learning applications. While standard integer-order neural ordinary differential equations (ODEs) have shown considerable success, they are limited in their capacity to model systems with memory effects and historical dependencies. Fractional calculus offers a mathematical framework capable of addressing this limitation, yet most current fractional neural networks use static memory weightings that cannot adapt to input-specific contextual requirements. This paper proposes a generalized neural Fractional Attention Differential Equation (FADE), which combines the memory-retention capabilities of fractional calculus with contextual learnable attention mechanisms. Our approach replaces fixed kernel functions in fractional operators with neural attention kernels that adaptively weight historical states based on their contextual relevance to current predictions. This allows our framework to selectively emphasize important temporal dependencies while filtering less relevant historical information. Our theoretical analysis establishes solution boundedness, problem well-posedness, and numerical equation solver convergence properties of the proposed model. Furthermore, through extensive evaluation on tasks such as fluid flow, graph learning problems and spatio-temporal traffic flow forecasting, we demonstrate that our adaptive attention-based fractional framework outperforms both integer-order neural ODE models and existing fractional approaches. The results confirm that our framework provides superior modeling capacity for complex dynamics with varying temporal dependencies. The code is available at \url{https://github.com/cuiwjTech/NeurIPS2025_FADE}.

Large Language Models (LLMs) capable of handling extended contexts are in high demand, yet their inference remains challenging due to substantial Key-Value (KV) cache size and high memory bandwidth requirements. Previous research has demonstrated that KV cache exhibits low-rank characteristics within the hidden dimension, suggesting the potential for effective compression. However, due to the widely adopted Rotary Position Embedding (RoPE) mechanism in modern LLMs, naive low‑-rank compression suffers severe accuracy degradation or creates a new speed bottleneck, as the low-rank cache must first be reconstructed in order to apply RoPE. In this paper, we introduce two key insights: first, the application of RoPE to the key vectors increases their variance, which in turn results in a higher rank; second, after the key vectors are transformed into the latent space, they largely maintain their representation across most layers. Based on these insights, we propose the Sparse Attention in Latent Space (SALS) framework. SALS projects the KV cache into a compact latent space via low-rank projection, and performs sparse token selection using RoPE-free query--key interactions in this space. By reconstructing only a small subset of important tokens, it avoids the overhead of full KV cache reconstruction. We comprehensively evaluate SALS on various tasks using two large-scale models: LLaMA2-7b-chat and Mistral-7b, and additionally verify its scalability on the RULER-128k benchmark with LLaMA3.1-8B-Instruct. Experimental results demonstrate that SALS achieves SOTA performance by maintaining competitive accuracy. Under different settings, SALS achieves 6.4-fold KV cache compression and 5.7-fold speed-up in the attention operator compared to FlashAttention2 on the 4K sequence. For the end-to-end throughput performance, we achieves 1.4-fold and 4.5-fold improvement compared to GPT-fast on 4k and 32K sequences, respectively. The source code will be publicly available in the future.

In this work, we investigate the challenging task of Generalized Category Discovery (GCD). Given datasets collected from open-world scenarios comprising both labeled and unlabeled images, GCD aims to classify all unlabeled images while simultaneously identifying unlabeled novel categories. The fundamental challenge in GCD tasks stems from inherent annotation discrepancies between seen and novel classes within the dataset. The lack of reliable label supervision for novel classes in unlabeled data leads to significant disparities in the model’s learning between old and novel classes, which is termed the bias risk. Recent advancements in GCD have employed the entropy maximization algorithm to alleviate the bias risk. However, they fail to provide debiased optimization for unlabeled data, leading to models that struggle with extracting discriminative features from such data. To address these challenges, we have created an Open-world pseudo-contrastive learning framework named OpcGCD. Our OpcGCD framework implements a dynamic category-wise threshold mechanism, which employs a parametric prototype classifiers to generate debiased pseudo-labels for unlabeled samples. To facilitate the learning of discriminative feature representations, our proposed OpcGCD employs debiased pseudo-labels in the formulation of a contrastive learning loss. Extensive evaluations conducted on multiple GCD benchmark datasets demonstrate the robustness and effectiveness of the approach.

Task-oriented point cloud sampling is a fundamental technique in 3D computer vision and has become a crucial step in numerous 3D applications. However, most state-of-the-art task-oriented sampling methods adopt a point-wise analysis strategy, making them susceptible to data redundancy. Taking inspiration from the abstract-to-detailed recognition process of the human visual system, we propose a novel voxel-to-point network framework called V2PNet for task-oriented point cloud sampling. Specifically, we first design a lightweight coarse-grained sampling module named Important Voxel Prediction (IMVP). This module adaptively outputs points from significant regions of the point cloud by explicitly modeling inter-region relationships, thereby reducing interference from redundant points. Then, the V2PNet framework seamlessly integrates the IMVP module with existing point-wise and task-oriented sampling networks, enabling joint training with downstream tasks. This creates a task-oriented coarse-to-fine-grained sampling pipeline that effectively samples representative and informative points from significant regions to represent the original point cloud. Moreover, to mitigate disturbances across similar regions, we introduce a voxel simplification loss function to enhance the discriminative voxel prediction. Extensive experiments demonstrate that V2PNet improves the performance of existing state-of-the-art task-oriented sampling models.

Considerable efforts have been devoted to addressing distributed backdoor attacks in federated learning (FL) systems. While significant progress has been made in enhancing the security of FL systems, our study reveals that there remains a false sense of security surrounding FL. We demonstrate that colluding malicious participants can effectively execute backdoor attacks during the FL training process, exhibiting high sparsity and stealthiness, which means they can evade common defense methods with only a few attack iterations. Our research highlights this vulnerability by proposing aCollusiveBackdoorAttack namedCoBA.CoBAis designed to enhance the sparsity and stealthiness of backdoor attacks by offering trigger tuning to facilitate learning of backdoor training data, controlling the bias of malicious local model updates, and applying the projected gradient descent technique. By conducting extensive empirical studies on 5 benchmark datasets, we make the following observations: 1)CoBAsuccessfully circumvents 15 state-of-the-art defense methods for robust FL; 2) Compared to existing backdoor attacks,CoBAconsistently achieves superior attack performance; and 3)CoBAcan achieve persistent poisoning effects through significantly sparse attack iterations. These findings raise substantial concerns regarding the integrity of FL and underscore the urgent need for heightened vigilance in defending against such attacks.

Video-based visible-infrared person re-identification (VVI-ReID) task focuses on cross-modality retrieval of pedestrian videos, which are captured in visible and infrared modalities by non-overlapping cameras across diverse scenes, and holds significant value for security surveillance scenarios. The challenges of this task mainly stem from three issues: the difficulty of capturing comprehensive spatio-temporal cues, intra-class variations within video sequences, and inter-modality discrepancies between visible and infrared data. Existing methods mainly try to address the modality gap or focus on one of the other aspects, but rarely do they jointly consider these key factors. Motivated by these core challenges, we propose the M3-ReID (Multi-View & Granularity & Modality) method, a unified framework that simultaneously enhances spatio-temporal feature extraction, intra-class discrimination, and cross-modality consistency. Specifically, to capture diverse spatio-temporal patterns, we design a Multi-View Learning module that leverages different spatial and temporal-spatial perspectives to adaptively emphasize diverse key regions and motion cues. To enhance intra-class modeling of each identity, we introduce a Multi-Granularity Representation strategy that optimizes features across both fine-grained frame level and coarse-grained video level by minimizing mutual information among redundant frames while enhancing identity representations. Furthermore, to bridge the visible-infrared gap, we propose a Multi-Modality Alignment mechanism that explicitly aligns metric learning and cross-modality matching goals, transforming features into a unified embedding space with modality consistency and class discrimination. Extensive experiments on benchmark VVI-ReID datasets demonstrate the superiority of our proposed M3-ReID framework against existing methods.

3D lane detection is a fundamental task in autonomous driving, and detecting 3D lanes from monocular images has been widely adopted due to the low computational cost and the property of lanes. Recent progresses have been made based on surrogate representations such as bird’s eye view (BEV) features. However, monocular BEV construction strictly relies on flat groud assumption, and the misalignment between perspective view and BEV is inevitable. In this paper, we propose Ada3DLane, a BEV-free, query based 3D lane detector, which adaptively generates queries with rich semantic and geometric information as well as lane interactions; adaptively samples perspective view image features in spatial and temporal domain; adaptively decode the sampled features for fast and accurate 3D lane detection. We conduct extensive experiments on benchmark lane detection datasets and outperforms previous state-of-the-art methods.

The rapid expansion and intelligent development of railway infrastructure are driving significant growth in railway engineering data. For the dispersed users across railway networks’ complex topology, traditional centralized storage systems are insufficient for their low-latency, cost-efficient data retrieval. Existing edge caching solutions based on multi-agent reinforcement learning fail to address the asymmetric relationships among railway nodes, such as data centers, stations, and sections, etc. Besides, the complexity of computing Nash equilibrium points also gets higher as the number of agents (edge caching servers) increases. This study introduces a Hierarchical Game model-based MADRL-driven Collaborative Edge Caching method(HG-MCEC) tailored for railway engineering data. By considering the distribution characteristics and caching strategy games among railway nodes, a hierarchical game model for collaborative edge caching is constructed. This model treats the railway edge caching as a multi-agent system, in which each railway node server is regarded as an agent. HG-MCEC utilizes deep learning to mitigate computational complexity and recognize agents’ asymmetry. Upper-level agents adjust cache replacement strategies according to environmental changes and decisionmaking experience. Lower-level agents, under the guidance of upper-level decisions, optimize collaborative caching strategies toward achieving hierarchical game equilibrium. Using a highspeed railway building information modeling data for validation, the method significantly outperforms existing approaches by enhancing content hit rates and reducing latency at edge caching servers while decreasing system content transmission costs.

Most previous learning-based graph matching algorithms solve the quadratic assignment problem (QAP) by dropping one or more of the matching constraints and adopting a relaxed assignment solver to obtain sub-optimal correspondences. Such relaxation may actually weaken the original graph matching problem, and in turn hurt the matching performance. In this paper, we propose a deep learning-based graph matching framework that works for the original QAP without compromising on the matching constraints. In particular, we design an affinityassignment prediction network to jointly learn the pairwise affinity and estimate the node assignments, and we then develop a differentiable solver inspired by the probabilistic perspective of the pairwise affinities. Aiming to obtain better matching results, the probabilistic solver refines the estimated assignments in an iterative manner to impose both discrete and one-to-one matching constraints. The proposed method is trained in a supervised manner, evaluated on several benchmarks related to semantic keypoint corresponding, matching of social networks and pure QAP instances. In all experiment, it exhibits state-of-the-art matching performance on all benchmarks.

Deploying quantized deep neural network (DNN) models with resource adaptation capabilities on ubiquitous Internet of Things (IoT) devices to provide high-quality AI services can leverage the benefits of compression and meet multi-scenario resource requirements. However, existing dynamic/mixed precision quantization requires retraining or special hardware, whereas post-training quantization (PTQ) has two limitations for resource adaptation: (i) The state-of-the-art PTQ methods only provide one fixed bitwidth model, which makes it challenging to adapt to the dynamic resources of IoT devices; (ii) Deploying multiple PTQ models with diverse bitwidths consumes large storage resources and switching overheads. To this end, this paper introduces a resource-friendly post-training integer-nesting quantization, i.e., NestQuant, for on-device quantized model switching on IoT devices. The proposed NestQuant incorporates the integer weight decomposition, which bit-wise splits quantized weights into higher-bit and lower-bit weights of integer data types. It also contains a decomposed weights nesting mechanism to optimize the higher-bit weights by adaptive rounding and nest them into the original quantized weights. In deployment, we can send and store only one NestQuant model and switch between the full-bit/part-bit model by paging in/out lower-bit weights to adapt to resource changes and reduce consumption. Experimental results on the ImageNet-1K pretrained DNNs demonstrated that the NestQuant model can achieve high performance in top-1 accuracy, and reduce in terms of data transmission, storage consumption, and switching overheads. In particular, the ResNet-101 with INT8 nesting INT6 can achieve 78.1% and 77.9% accuracy for full-bit and part-bit models, respectively, and reduce switching overheads by approximately 78.1% compared with diverse bitwidths PTQ models. Code:https://github.com/jianhayes/NESTQUANT.

Downsampling is a crucial task for processing large scale and/or dense point clouds with limited resources. Owing to the development of deep learning, approaches of task-oriented point cloud downsampling have significant performance gains in preserving geometric information. However, most downsamling methods are limited by the disordered and unstructured point cloud data, making it difficult to continually improve the performance. To address this issue, we propose a light-weight Transformer network (LighTN) for the task-oriented point cloud downsampling as an end-to-end solution. In LighTN, we design an energy-efficient and permutation invariant single-head self-correlation module to extract refined global geometric features. Moreover, we present a novel sampling loss function to guide LighTN to focus on critical point cloud regions with more uniform distributions and prominent point coverage. Extensive experiments on classification, registration, and reconstruction tasks demonstrate that LighTN can achieve the state-of-the-art performance-overhead tradeoff and high-quality qualitative results.

Point-voxel 3D object detectors have achieved impressive performance in complex traffic scenes. However, they utilize the 3D sparse convolution (spconv) layers with fixed receptive fields, such as voxel-based detectors, and inherit the fixed sphere radius from point-based methods for generating the features of keypoints, which make them weak in adaptively modeling various geometrical deformations and sizes of real objects. To tackle this issue, we propose a shape-adaptive set abstraction network (SASAN) for point-voxel 3D object detection. First, the proposal and offset generation module is adopted to learn the coordinates and confidences of 3D proposals and shape-adaptive offsets of the certain number of offset points for each voxel. Meanwhile, an extra offset supervision task is employed to guide the learning of shifting values of offset points, aiming at motivating the predicted offsets to preferably adapt to the various shapes of objects. Then, the shape-adaptive set abstraction module is proposed to extract multiscale keypoints features by grouping the neighboring offset points' features, as well as features learned from adjacent raw points and the 2-D bird-view map. Finally, the region of interest (RoI)-grid proposal refinement module is used to aggregate the keypoints features for further proposal refinement and confidence prediction. Extensive experiments on the competitive KITTI 3D detection benchmark demonstrate that the proposed SASAN gains superior performance as compared with state-of-the-art methods.

Federated recommender systems (FRSs), with their improved privacy-preserving advantages to jointly train recommendation models from numerous devices while keeping user data distributed, have been widely explored in modern recommender systems (RSs). However, conventional FRSs require transmitting the entire model between the server and clients, which brings a huge carbon footprint for cost-conscious cross-device learning tasks. While several efforts have been dedicated to improving the efficiency of FRSs, it's suboptimal to treat the whole model as the objective of compact design. Besides, current research fails to handle the out-of-vocabulary (OOV) issue in real-world FRSs, where the items only occasionally appear in the testing phase but were not observed during the training process, which is another practical challenge and has not been well studied yet. To this end, we propose a privacy-enhanced federated recommendation framework with shared hash embedding, PrivFR, in cross-device settings, which is an efficient representation mechanism specialized for the embedding parameters without compromising the model capability. Specifically, it represents items in a resource-efficient way by delicately utilizing shared hash embedding and multiple hash functions. As such, it just maintains a small shared pool of hash embedding in local clients, rather than fitting all embedding vectors for each item, which can exactly achieve the dual advantages of conserving resources and handling the OOV issue. What's more, we prove that this mechanism can protect the data privacy of local clients from a theoretical perspective. Extensive experiments show that our method not only effectively reduces storage and communication overheads, but also outperforms state-of-the-art FRSs.

Deploying the convolutional neural network (CNN) model on Internet of Things (IoT) devices to provide diverse recognition services has received increasing attention. Due to the limited storage, computing, and other resources of IoT devices, it has become mainstream to first train the CNN model on the edge/cloud server and then send the trained CNN to the IoT device. However, most existing related methods suffer from two limitations, (i) low performance due to service interference or insufficient mutual assistance, and (ii) large memory resources and switching resource overhead. To this end, this article proposes a resource-efficient multiple recognition services framework for IoT devices. The proposed framework is based on the edge server-assisted IoT device training of the CNN model, and the framework includes a deeper weight adaptation (DeepWAdapt) algorithm to mitigate service interference. The DeepWAdapt algorithm consists of a set of learnable masks, and by inserting these masks into the appropriate layers of the CNN model, it mitigates mutual interference between services caused by training a single CNN model for multiple services. Each service has a specific set of masks. These learnable masks work like keys for each service, selecting appropriate and specific features for each service from a shared feature set. Experimental results demonstrate that the DeepWAdapt outperforms other state-of-the-art methods on image-level classification services and pixel-level dense prediction services. Specifically, when executing 40 services based on ResNet18, the proposed DeepWAdapt achieves 66.82% F1-score on the CelebA dataset, which is +2.61% F1-score than the previous state-of-the-art result. In addition, compared with the routing method, our proposed DeepWAdapt also reduces network transmission traffic by approximately 35%.

Cloud-assisted Internet of Things (IoT) device deployment of deep neural networks (DNNs) promotes On-device deep learning to provide users with ubiquitous high-quality services by solving the contradiction between insufficient IoT device resources and intensive demand for high-performance DNN resources. However, most existing methods optimize DNNs by considering one or two terms of transmission, computation, and storage resources, but do not consider all three terms at the same time in cloud-assisted IoT device deployment and updating DNNs. To this end, we propose a non-learnable module-based ResNet and a cloud-assisted on-device deep learning framework, ReFrame, based on the consideration of three indicators: model transmission parameters, computation resources, and storage resources. In the proposed method, we first specify that some parameters in DNNs are non-learnable and randomly initialized, so that, these parameters can be saved and reproduced with a few random seeds. By doing so, the cloud only transmits random seeds and learnable parameters to reduce the number of parameter transmissions. Second, we reduce the computation resource consumption of the model by introducing computation-friendly operators, such as pooling, to replace vanilla convolutions. Finally, since random seeds are used to save non-learnable model parameters, on IoT devices we only need to store random seeds and learnable parameters to reproduce the well-trained model. Compared with saving the complete model, our method greatly reduces IoT device storage resource consumption. Experimental results on image classification, object detection, and semantic segmentation tasks demonstrate the effectiveness of the proposed method. Specifically, on the CIFAR-10, our proposed method reduces approximately 89% of FLOPs and 90% of transmitted data in the prototype system compared to ResNet-18.

Deploying a single deep neural network (DNN) model on ubiquitous Internet of Things (IoT) devices to provide multiple vision services (e.g., semantic segmentation service, facial attribute recognition service) has attracted significant interest from industry and academia. However, the most relevant studies have two limitations: (i) the multiple services DNN commonly has low performance due to the gradient interference between different vision services; (ii) the multiple services DNN has low inference speed on IoT devices. To this end, this paper introduces a multiple vision services acceleration framework for IoT devices. The proposed MCBNet contains a novel multi-controllable branching structure to control different services explicitly for improving multiple services performance and a service balance layer with the per-channel trainable masks in specific branches for mitigating gradient interference. In deployment, the proposed framework converts the multi-branch topology MCBNet model into the plain topology Rep-MCBNet model by re-parameterization technique for accelerating services inference speed. Experimental results on the CelebA and Cityscapes datasets show that our proposed framework outperforms the state-of-the-art approaches in terms of performance and inference speed. In particular, we deploy the Rep-MCBNet on the laptop with Intel mobile processors demonstrating that the proposed framework achieves 1.9× faster processing frames per second than ETR-NLP, with 1.5× fewer parameters and 1.3× fewer floating-point operations.

In recent years, decision-making research based on Deep Learning has gained significant attention, particularly in the domain of urban traffic flow prediction. This paper addresses the critical need for enhanced accuracy and speed in predicting traffic flow trends by utilizing state-of-the-art time series prediction algorithms. This paper propose a lightweight but efficient linear-based method specifically designed to optimize time series forecasting for traffic flow scenarios. Independent-channel and patch time series in our method ensures high prediction accuracy while maintaining a balance between computational efficiency and processing speed. By integrating the power of deep learning with the robustness of traditional statistical methods, the linear-based method effectively manages large-scale traffic prediction tasks, including both univariate and multivariate time series forecasting through independent channels and patch series. A comprehensive series of experiments demonstrates the superior performance of our algorithm in large-scale traffic flow prediction, comparing it favorably with other time series prediction algorithms. This study aims to introduce significant advancements in traffic flow prediction, providing valuable decision-making tools for community officials and management departments.Additionally, after testing proposed method on non-traffic datasets, it also achieves similarly remarkable results. Across various experiments, our method consistently delivers faster, more accurate, and more efficient predictions, surpassing the performance of several other models.

The Unaligned Multi-view Clustering (UMC) problem is currently receiving widespread attention, focusing on clustering unaligned multi-view data generated in real-world applications. Although some algorithms have emerged to address this issue, there still exist the following drawbacks: 1) The fully unknown correspondence of samples across views can significantly limit the exploration of consistent clustering structure. 2) The fixed representation space makes it difficult to mine the comprehensive information in the original data. 3) Unbiased tensor rank approximation is desired to capture the high-order correlation among different views. To address these issues, we proposed a novel UMC framework termed Tensorized Unaligned Multi-view Clustering with Multi-scale Representation Learning (TUMCR). Specifically, TUMCR designs a multi-scale representation learning and alignment framework, which constructs multi-scale representation spaces to comprehensively explore the unknown correspondence across views. Then, a tensorial multi-scale fusion module is proposed to fuse multi-scale representations and explore the high-order correlation hidden in different views, which utilizes the Enhanced Tensor Rank (ETR) to learn the low-rank structure. Furthermore, TUMCR is solved by an efficient algorithm with good convergence. Extensive experiments on different types of datasets demonstrate the effectiveness and superiority of our TUMCR compared with state-of-the-art methods. Our code is publicly available at: https://github.com/jijintian/TUMCR.

Few-shot 3D point cloud semantic segmentation aims to segment query point clouds with only a few annotated support point clouds. Existing prototype-based methods learn prototypes from the 3D support set to guide the segmentation of query point clouds. However, they encounter the challenge of low prototype quality due to constrained semantic information in the 3D support set and class information bias between support and query sets. To address these issues, in this paper, we propose a novel framework called Generated and Pseudo Content guided Prototype Refinement (GPCPR), which explicitly leverages LLM-generated content and reliable query context to enhance prototype quality. GPCPR achieves prototype refinement through two core components: LLM-driven Generated Content-guided Prototype Refinement (GCPR) and Pseudo Query Context-guided Prototype Refinement (PCPR). Specifically, GCPR integrates diverse and differentiated class descriptions generated by large language models to enrich prototypes with comprehensive semantic knowledge. PCPR further aggregates reliable class-specific pseudo-query context to mitigate class information bias and generate more suitable query-specific prototypes. Furthermore, we introduce a dual-distillation regularization term, enabling knowledge transfer between early-stage entities (prototypes or pseudo predictions) and their deeper counterparts to enhance refinement. Extensive experiments demonstrate the superiority of our method, surpassing the state-of-the-art methods by up to 12.10% and 13.75% mIoU on S3DIS and ScanNet, respectively.

Graph neural networks (GNNs) are recognized for their strong performance across various applications, with the backpropagation (BP) algorithm playing a central role in the development of most GNN models. However, despite its effectiveness, BP has limitations that challenge its biological plausibility and affect the efficiency, scalability and parallelism of training neural networks for graph-based tasks. While several non-backpropagation (non-BP) training algorithms, such as the direct feedback alignment (DFA), have been successfully applied to fully-connected and convolutional network components for handling Euclidean data, directly adapting these non-BP frameworks to manage non-Euclidean graph data in GNN models presents significant challenges. These challenges primarily arise from the violation of the independent and identically distributed (i.i.d.) assumption in graph data and the difficulty in accessing prediction errors for all samples (nodes) within the graph. To overcome these obstacles, in this paper we propose DFA-GNN, a novel forward learning framework tailored for GNNs with a case study of semi-supervised learning. The proposed method breaks the limitations of BP by using a dedicated forward training mechanism. Specifically, DFA-GNN extends the principles of DFA to adapt to graph data and unique architecture of GNNs, which incorporates the information of graph topology into the feedback links to accommodate the non-Euclidean characteristics of graph data. Additionally, for semi-supervised graph learning tasks, we developed a pseudo error generator that spreads residual errors from training data to create a pseudo error for each unlabeled node. These pseudo errors are then utilized to train GNNs using DFA. Extensive experiments on 10 public benchmarks reveal that our learning framework outperforms not only previous non-BP methods but also the standard BP methods, and it exhibits excellent robustness against various types of noise and attacks.

Task-oriented point cloud sampling methods have attracted considerable attention for their ability to adaptively select important point sets based on downstream tasks, achieving an excellent balance between data simplification and task performance. However, existing task-oriented sampling models, primarily based on single-branch designs, struggle to fully extract features from input point clouds that comprehensively reflect multi-dimensional key information, thus limiting their sampling performance. In this paper, we introduce a dual-branch sampling network, named DBS-NET, which conducts crucial point sampling from both the global and local importance perspectives separately before merging them, thereby preserving multi-dimensional key information of the input data during the sampling process. Qualitative and quantitative experimental results demonstrate the competitive performance of DBS-NET on the classification benchmark task.

Abstract Ridesharing is an effective approach to alleviate traffic congestion. In most existing works, drivers and passengers are assigned prices without considering the constraints of desired benefits. This paper investigates ridesharing by formulating a matching and pricing problem to maximize the total payoff of drivers, with the constraints of desired benefit and quality of experience. An efficient algorithm is proposed to solve the formulated problem based on coalitional double auction. Secondary pricing based strategy and sacrificed minimum bid based strategy are proposed to support the algorithm. This paper also proves that the proposed algorithm can achieve a Nash-stable coalition partition in finite steps, and the proposed two strategies guarantee truthfulness, individually rational and budget balance. Extensive simulation results on the real-world dataset of taxi trajectory in Beijing city show that the proposed algorithm outperforms the existing ones, in terms of average total payoff of drivers while meeting the benefits of passengers.

In Internet of Vehicles (IoV), it is critical to collect sufficient data for model training, to support vehicular intelligent applications. However, the environment of IoV is highly dynamic due to the mobility of vehicles, making it challenging to efficiently allocate resources for data collection. Additionally, timely training of machine learning models with collected data is important for accurate representation in a constantly changing environment. This article aims to improve the performance of model training by collecting sufficient data from vehicles in IoV. A system throughput maximization problem is formulated under the limited bandwidth, storage, and computing resources, which is an NP-hard problem. To solve the problem, an iterative algorithm, namely, the iterative algorithm based on approaching minimum bandwidth (IAMB), is proposed to preferentially collect data from the vehicles with sufficient data and reliable communication quality. Besides, a genetic algorithm, namely, the genetic algorithm based on approaching minimum bandwidth (GAMB), is proposed to further improve the probability of superior individual by replacing operation. We also customize three greedy strategy-based algorithms as the baselines. Extensive experimental results show that our proposed algorithms outperform the baseline algorithms for all cases. Specifically, IAMB and GAMB can improve the throughput by up to 5% and 8%, respectively, compared with baseline algorithms. In addition, the customized genetic algorithm is also superior to the iterative algorithm on performance of system throughput. Moreover, the customized genetic algorithm is more stable than the proposed iterative algorithm in terms of system throughput for model training in the dynamic network environment.

Visible-Infrared person Re-IDentification (VI-ReID) is a challenging cross-modality image retrieval task that aims to match pedestrians’ images across visible and infrared cameras. To solve the modality gap, existing mainstream methods adopt a learning paradigm converting the image retrieval task into an image classification task with cross-entropy loss and auxiliary metric learning losses. These losses follow the strategy of adjusting the distribution of extracted embeddings to reduce the intra-class distance and increase the inter-class distance. However, such objectives do not precisely correspond to the final test setting of the retrieval task, resulting in a new gap at the optimization level. By rethinking these keys of VI-ReID, we propose a simple and effective method, the Multi-level Cross-modality Joint Alignment (MCJA), bridging both the modality and objective-level gap. For the former, we design the Visible-Infrared Modality Coordinator in the image space and propose the Modality Distribution Adapter in the feature space, effectively reducing modality discrepancy of the feature extraction process. For the latter, we introduce a new Cross-Modality Retrieval loss. It is the first work to constrain from the perspective of the ranking list in the VI-ReID, aligning with the goal of the testing stage. Moreover, to strengthen the robustness and cross-modality retrieval ability, we further introduce a Multi-Spectral Enhanced Ranking strategy for the testing phase. Based on the global feature only, our method outperforms existing methods by a large margin, achieving the remarkable rank-1 of 89.51% and mAP of 87.58% on the most challenging single-shot setting and all-search mode of the SYSU-MM01 dataset.

Cross-modality person re-identification task is a challenging task aiming to recognize images of the same identity between different modalities. To alleviate the cross-modality discrepancies between images, existing approaches mainly guide models to mine modality invariant features. Although those approaches are effective, they lose the modality-specific features that include important information beneficial to VI-ReID. Therefore, some approaches are using generative adversarial networks to compensate for modality information. However, the quality of images generated by these methods is usually poor, and most of them focus only on the learning of modality-sharable features. To solve these problems, this paper proposes a generative-based cross-modality image fusion strategy (GC-IFS), which can generate high-quality cross-modality paired images and fuse the information of the two modalities. Firstly, considering the importance of the identity discriminative information of the generated image, we propose a contrastive-learning image generation (CLIG) network to generate cross-modality paired images. Meanwhile, to fully integrate and utilize the information of the two modalities and eliminate the influence of cross-modality discrepancies, we design a part-based dual multi-modality feature fusion (P-DMFF) module to extract the unified feature representation. Extensive experiments on SYSU-MM01 and RegDB datasets demonstrate that our strategy outperforms the state-of-the-art methods for the VI-ReID task.

The Visible-Infrared (VIS-IR) object detection is a challenging detection task, which combines visible and infrared data to give information on the category and location of objects in the scene. Therefore, the core of this task is to combine complementary information in the visible and infrared modalities to provide more object detection results for detection. The existing methods mainly face the problem of insufficient ability to perceive and combine visible-infrared modal information and have difficulty in balancing the optimization directions of the fusion and detection tasks. To solve these problem, we propose the MMI-Det which is a multi-modal fusion method for visible and infrared object detection. The method can provide a good combination of complementary information in the visible-infrared modalities and output accurate and robust object information. Specifically, to improve the ability of the model to perceive environment at the visible-infrared image level, we designed the Contour Enhancement Module. Furthermore, to extract complementary information from VIS and IR modalities, we design the Fusion Focus Module. It can extract different frequency spectral features of the visible and infrared modalities and focus on the key information of the object at different spatial locations. Moreover, we design the Contrast Bridge Module to improve the ability to extract modal invariant features in the visible-infrared scene. Finally, to ensure that our model can balance the optimization directions of image fusion and object detection, we design the Info Guided Module as a way to improve the effectiveness of the model’s training optimization. We implement extensive experiments on the public FLIR, M3FD, LLVIP, TNO and MSRS datasets, and compared with previous methods, our method achieves better performance with powerful multi-modal information perception capabilities.

Object Re-identification (Re-ID), which includes person Re-ID and vehicle Re-ID, is one of the core technologies of the intelligent transportation system. Existing supervised Re-ID studies mainly focus on discriminative feature learning (e.g., attention-based methods) or metric learning (e.g., triplet-loss-based methods) to obtain more accurate matches between the probe object and the positive gallery. However, they both pay less attention to global structure information (GSI) buried in the overall datasets. In this paper, we go beyond the traditional methods that are either unaware of or locally perceiving to GSI, and consider exploring the structural relationships among all the object instances of a dataset via a graph. Specifically, we construct a graph across the entire dataset, where each object instance is treated as a node and edges are assigned with the help of a classic algorithm like KNN. Seeing that a binary edge label can be used to predict whether its associated nodes belong to the same identity, we naturally formulate the problem of Re-ID as a new link prediction problem. Inspired by the superior capacity of capturing structure information of graph convolutional networks (GCN), a GCN-based global structure embedded network (GSE-Net) is proposed to take the graph as input and output a set of linkage likelihoods. During testing, we perform the evaluation according to the node features or estimated linkage likelihood via a graph where nodes include query and gallery images. Extensive experiments demonstrate that our proposed method outperforms the state-of-the-arts on both person and vehicle Re-ID benchmarks.

Unsupervised domain adaptive object detection methods aim to transfer knowledge from the label-sufficient domain to the unlabeled domain. Most existing works minimize domain disparity by concentrating on different levels through adversarial learning. However, adversarial learning do not consider the different influences on under-aligned and well-aligned samples as they merely match distinct distributions with consistent weight. To address this issue, we design a novel scale-disentangled and uncertainty-guided alignment for domain-adaptive object detection (SDUGA), consisting of three main components: (1) Disentangled scale coarse module, which decouples scale information from global image features and performs individual alignment across domains for the corresponding scale by training domain classifiers in an adversarial learning manner; (2) Disentangled scale fine module, which generalizes the disentangled scale alignment to instance-level adaptation, reinforcing the distribution alignment across domains from multi-scale local instance level; (3) Uncertainty-guided coarse-to-fine attention alignment, which adjusts weights for various samples adaptively by generating the uncertainty-guided attention map, thus enforcing the detector to converge more on alignment for under-aligned samples and avoid misaligning well-aligned ones. Extensive experiments over three challenging domain-shift object detection scenarios demonstrate that SDUGA gains superior performance compared to state-of-the-art methods.

Extracting features from image data on Internet of Things (IoT) devices to reduce the amount of data that needs to be uploaded to cloud/edge servers has received increasing attention. However, most of the existing related approaches suffer from two major limitations, (i) low performance and high network traffic, and (ii) a lot of storage resource consumption. To this end, we propose a resource-efficient feature extraction framework for image processing in IoT devices. The proposed framework consists of the edge-assisted extractor generation method and the NestE method. The extractor generated by the edge-assisted extractor generation method can extract the features required by the application, which can not only avoid the IoT device uploading useless feature data but also improve application performance. The proposed NestE generates a nonredundant subextractor by splitting the extractor into multiple subextractors, removing redundant subextractors, and nesting small-capacity subextractors in large-capacity subextractors in a parameter-sharing manner. Compared with deploying multiple independent subextractors on IoT devices, deploying the nonredundant multifunctional extractor can save considerable storage resources and switching overhead. Extensive experimental results show that the proposed framework reduces the storage footprint by approximately 90.7% and switching overhead by approximately 92.4% compared with deploying independent subextractors when using the classical principal component analysis algorithm.

The deployment of Deep Neural Networks (DNNs) requires significant computational and storage resources, which is challenging for resource-constrained end devices. To this end, collaborative deep inference is proposed, in which the DNN is divided into two parts and executed on the end device and cloud respectively. The selection of DNN partition point is the key challenge to realize end-cloud collaborative deep inference, especially in mobile environments with unstable networks. In this paper, we propose a Real-time Adaptive Partition (RAP) framework, in which a fast split point decision algorithm is proposed to realize real-time adaptive DNN model partition in the mobile network. A weighted joint optimization of DNN quantization loss, inference and transmission latency is performed. We further propose a Joint Multi-user Model Partition and Resource Allocation (JM-MPRA) algorithm under RAP framework. JM-MPRA aims to guarantee the optimized latency, accuracy and resource utilization in the multi-user scene. Experimental evaluations have demonstrated the effectiveness of RAP with JM-MPRA in improving the performance of real-time end-cloud collaborative inference in both stable and unstable mobile networks. Compared with the state-of-the-art methods, the proposed approaches can achieve up to 5.06x decrease in inference latency and bring performance improvement of 1.52% in inference accuracy.

Graph convolutional networks (GCNs) are widely believed to perform well in the graph node classification task, and homophily assumption plays a core rule in the design of previous GCNs. However, some recent advances on this area have pointed out that homophily may not be a necessity for GCNs. For deeper analysis of the critical factor affecting the performance of GCNs, we first propose a metric, namely, neighborhood class consistency (NCC), to quantitatively characterize the neighborhood patterns of graph datasets. Experiments surprisingly illustrate that our NCC is a better indicator, in comparison to the widely used homophily metrics, to estimate GCN performance for node classification. Furthermore, we propose a topology augmentation graph convolutional network (TA-GCN) framework under the guidance of the NCC metric, which simultaneously learns an augmented graph topology with higher NCC score and a node classifier based on the augmented graph topology. Extensive experiments on six public benchmarks clearly show that the proposed TA-GCN derives ideal topology with higher NCC score given the original graph topology and raw features, and it achieves excellent performance for semi-supervised node classification in comparison to several state-of-the-art (SOTA) baseline algorithms.

With the advancement of intelligent and networked technology, the Industrial Internet of Things (IIoT) faces an escalating threat from cyberattacks, especially by Advanced Persistent Threat (APT) attacks. These novel and complex attacks, characterized by their dynamic nature and life-long duration, pose significant challenges to existing security protection methods. The challenges are twofold, i.e., sparse reward problem in the long-lasting attack, and partial observation of attack actions. To this end, we propose a Security-as-a-Service based reinforcement learning method, namely Attention Augmented Dueling Deep Q-learning Network (AD2QN), to make real-time defense strategies for the hot standby IIoT. First, we build the attack-defend confrontation model as black boxes interact with the IIoT environment to play a long-lasting partially observable zero-sum stochastic game on the server. Then, to dynamically generate optimal defense strategies as the service, AD2QN is proposed employing information completion and prediction to more informed action selection. Furthermore, AD2QN utilizes an iteratively updated reward network to deal with the sparse reward problem. Extensive simulation results shown that the defense strategies generated by our method have a higher defense success rate and a stable defense performance with the average success rate of 0.7384, while the average success rate of baseline methods was 0.7375, in the best case.

Visual Transformers (ViTs) and Convolutional Neural Networks (CNNs) are the two primary backbone structures extensively used in various vision tasks. Generating transferable adversarial examples for ViTs is difficult due to ViTs’ superior robustness, while transferring adversarial examples across ViTs and CNNs is even harder, since their structures and mechanisms for processing images are fundamentally distinct. In this work, we propose a novel attack method named Momentum Integrated Gradients (MIG), which not only attacks ViTs with high success rate, but also exhibits impressive transferability across ViTs and CNNs. Specifically, we use integrated gradients rather than gradients to steer the generation of adversarial perturbations, inspired by the observation that integrated gradients of images demonstrate higher similarity across models in comparison to regular gradients. Then we acquire the accumulated gradients by combining the integrated gradients from previous iterations with the current ones in a momentum manner and use their sign to modify the perturbations iteratively. We conduct extensive experiments to demonstrate that adversarial examples obtained using MIG show stronger transferability, resulting in significant improvements over state-of-the-art methods for both CNN and ViT models.

Deploying high-performance convolutional neural networks (CNNs) on ubiquitous Internet of Things (IoT) devices to provide convenient services has attracted increasing attention. However, most existing studies have two limitations, (i) low performance due to lack of effective learning strategies; (ii) difficulty to measure resource consumption due to lack of real deployment. To this end, this paper proposes a novel localized knowledge distillation (LKD) method to train the resource-efficient CNN and implements a cloud-assisted system to evaluate the on-device performance. The proposed LKD follows the layer-wise heterogeneous information distribution in the CNN and distills the knowledge from features that contains the most crucial knowledge to the resource-efficient CNN. Thus, the knowledge guided to learn by limited parameters is reduced to the crucial part, which is more reliable for the resource-efficient CNN. The cloud-assisted system consists cloud server and IoT devices, which allows the training and deployment of the resource-efficient CNN, and the measurement of the corresponding resource consumption. Experimental results show that the proposed LKD could improve the performance on standard benchmarks close to the counterpart CNN, and the robustness on corrupted data by approximately 11.3% for the resource-efficient CNN. The measurements on the cloud-assisted system also demonstrate the resource efficiency for transmission and on-device running.

Federated learning (FL) is a privacy-aware computing framework that enables multiple clients to collaborate in solving machine learning problems. In real scenarios, non-IID data held by different edge devices will degrade the performance of global FL models. To address this issue, most FL methods utilize cluster algorithms to group clients with similar distributions. However, these methods do not fully utilize the distribution features of client data, resulting in a lack of generalization in the cluster model. In order to make the cluster more suitable for the distribution features of user data, we propose a clustering-ensemble based federated learning method (FedCE) that sets each client associated with multiple clusters. We extract the features of client distributions to quantify the relationship between clients and clusters, and optimize the local model of clients through the historical performance of the cluster model. Furthermore, we dynamically estimate the number of clusters each client belongs to through the diversity of client performance. We conduct experiments on scenarios with mixture two-distributions, three-distributions and Dirichlet-distributions. The results show that the FedCE algorithm has better performance than the state-of-the-art clustered FL methods in both cluster and client models under different data distributions.

Collaborative 3D object detection, with its improved interaction advantage among multiple agents, has been widely explored in autonomous driving. However, existing collaborative 3D object detectors in a fully supervised paradigm heavily rely on large-scale annotated 3D bounding boxes, which is labor-intensive and time-consuming. To tackle this issue, we propose a sparsely supervised collaborative 3D object detection framework SSC3OD, which only requires each agent to randomly label one object in the scene. Specifically, this model consists of two novel components, i.e., the pillar-based masked autoencoder (Pillar-MAE) and the instance mining module. The Pillar-MAE module aims to reason over high-level semantics in a self-supervised manner, and the instance mining module generates high-quality pseudo labels for collaborative detectors online. By introducing these simple yet effective mechanisms, the proposed SSC3OD can alleviate the adverse impacts of incomplete annotations. We generate sparse labels based on collaborative perception datasets to evaluate our method. Extensive experiments on three large-scale datasets reveal that our proposed SSC3OD can effectively improve the performance of sparsely supervised collaborative 3D object detectors.

Abstract Parameter servers (PSs) placement is one of the most important factors for global model training on distributed deep learning. This paper formulates a novel problem for placement strategy of PSs in the dynamic available storage capacity, with the objective of minimizing the training time of the distributed deep learning under the constraints of storage capacity and the number of local PSs. Then, we provide the proof for the NP-hardness of the proposed problem. The whole training epochs are divided into two parts, i.e. the first epoch and the other epochs. For the first epoch, an approximation algorithm and a rounding algorithm are proposed in this paper, to solve the proposed problem. For the other epochs, an adjustment algorithm is proposed, by continuously adjusting the decisions for placement strategy of PSs to decrease the training time of the global model. Simulation results show that the proposed approximation algorithm and rounding algorithm perform better than existing works for all cases, in terms of the training time of global model. Meanwhile, the training time of global model for the proposed approximation algorithm is very close to that for optimal solution generated by the brute-force approach for all cases. Besides, the integrated algorithm outperforms the existing works when the available storage capacity varies during the training.

Summary Access control is an important technique in information security that allows legitimate users to gain access to and prevent unauthorized users from getting access to resources in a system. The restriction between access from a user and a shared file of the data owner can be determined by the access policy. In most existing access control models, it is assumed that all entities, including users, the third party, and cloud service provider (CSP), are in the same trust domain. However, in cloud computing environments, it is usually assumed that the CSP cannot be fully trusted, and the data owner (DO) is desired to have the absolute initiative to control data access. This article proposes a blockchain‐based access control scheme for cloud computing, in which the DO maintains an access matrix to describe the access policy. Then, the public keys of all nodes and the access matrix are stored in the blockchain, to ensure the security of the proposed scheme. The DOs can encrypt the large shared files once using a symmetric key in a long time. And they also can encrypt the symmetric key in parallel using the public key of authorized users in a short time. Security analysis proves that the proposed scheme is able to prevent outsourced files from unauthorized access and collusion attack. Experimental results show that the proposed scheme outperforms the existing baselines in terms of overheads on computation and storage. On average, the computation overhead of the proposed scheme is lower than that of the scheme SVPAC, PpBAC, and Timely CP‐ABE by 25.37%, 45.46%, and 36.44%, respectively. The communication overhead of the proposed scheme is lower than that of the scheme Timely CP‐ABE by 17.16%, and it is more secure, although it is higher than that of the scheme SVPAC and PpBAC by 5.88% and 39.05%. And the storage overhead of the proposed scheme is lower than that of the scheme SVPAC, PpBAC, and Timely CP‐ABE by 59.36%, 20.25%, and 61.88%, respectively.

Abstract As a critical component of ensuring the safe and stable operation of trains, the detection of bird’s nests on the rail catenary has always been essential. Low-resolution images and the lack of labelled data, however, make it difficult to detect smaller bird’s nests (those occupying small pixels in the input image). Previous solution relies on manual online patrol or offline video playback, which severely limits the detection efficiency. Previously, this challenge was addressed by manual online patrol or offline video playback, which severely limits detection efficiency. We propose in this work a context-guided coarse-to-fine detection model (CG-CFDM) for solving the bird’s nest detection problem. This solution consists of a context reasoning module and a coarse-to-fine detection network. By detecting domains and matching templates, the context reasoning module generates new labelled context bounding boxes, thereby reducing the burden of annotation. As a result of its delicately designed architecture and powerful representation learning ability, this trained coarse-to-fine detection network further facilitates the detection of bird’s nests in an efficient and accurate manner. Extensive experiments demonstrate that the proposed approach is superior to existing methods in terms of performance and has a great deal of potential for detecting bird’s nests.

Partial multilabel learning (PML) aims to learn from training data, where each instance is associated with a set of candidate labels, among which only a part is correct. The common strategy to deal with such a problem is disambiguation, that is, identifying the ground-truth labels from the given candidate labels. However, the existing PML approaches always focus on leveraging the instance relationship to disambiguate the given noisy label space, while the potentially useful information in label space is not effectively explored. Meanwhile, the existence of noise and outliers in training data also makes the disambiguation operation less reliable, which inevitably decreases the robustness of the learned model. In this article, we propose a prior label knowledge regularized self-representation PML approach, called PAKS, where the self-representation scheme and prior label knowledge are jointly incorporated into a unified framework. Specifically, we introduce a self-representation model with a low-rank constraint, which aims to learn the subspace representations of distinct instances and explore the high-order underlying correlation among different instances. Meanwhile, we incorporate prior label knowledge into the above self-representation model, where the prior label knowledge is regarded as the complement of features to obtain an accurate self-representation matrix. The core of PAKS is to take advantage of the data membership preference, which is derived from the prior label knowledge, to purify the discovered membership of the data and accordingly obtain more representative feature subspace for model induction. Enormous experiments on both synthetic and real-world datasets show that our proposed approach can achieve superior or comparable performance to state-of-the-art approaches.

Deploying deep convolutional neural network (CNN) models on ubiquitous Internet of Things (IoT) devices has attracted much attention from industry and academia since it greatly facilitates our lives by providing various rapid-response services. Due to the limited resources of IoT devices, cloud-assisted training of CNN models has become the mainstream. However, most existing related works suffer froma large amount of model parameter transmission and weak model robustness. To this end, this paper proposes a cloud-assisted CNN training framework with low model parameter transmission and strong model robustness. In the proposed framework, we first introduce MonoCNN, which contains only a few learnable filters, and other filters are nonlearnable. These nonlearnable filter parameters are generated according to certain rules, i.e., the filter generation function (FGF), and can be saved and reproduced by a few random seeds. Thus, the cloud server only needs to send these learnable filters and a few seeds to the IoT device. Compared to transmitting all model parameters, sending several learnable filter parameters and seeds can significantly reduce parameter transmission. Then, we investigate multiple FGFs and enable the IoT device to use the FGF to generate multiple filters and combine them into MonoCNN. Thus, MonoCNN is affected not only by the training data but also by the FGF. The rules of the FGF play a role in regularizing the MonoCNN, thereby improving its robustness. Experimental results show that compared to state-of-the-art methods, our proposed framework can reduce a large amount of model parameter transfer between the cloud server and the IoT device while improving the performance by approximately 2.2% when dealing with corrupted data.

In this work, we focus on Interactive Human Parsing (IHP), which aims to segment a human image into multiple human body parts with guidance from users’ interactions. This new task inherits the class-aware property of human parsing, which cannot be well solved by traditional interactive image segmentation approaches that are generally class-agnostic. To tackle this new task, we first exploit user clicks to identify different human parts in the given image. These clicks are subsequently transformed into semantic-aware localization maps, which are concatenated with the RGB image to form the input of the segmentation network and generate the initial parsing result. To enable the network to better perceive user's purpose during the correction process, we investigate several principal ways for the refinement, and reveal that random-sampling-based click augmentation is the best way for promoting the correction effectiveness. Furthermore, we also propose a semantic-perceiving loss (SP-loss) to augment the training, which can effectively exploit the semantic relationships of clicks for better optimization. To the best knowledge, this work is the first attempt to tackle the human parsing task under the interactive setting. Our IHP solution achieves 85% mIoU on the benchmark LIP, 80% mIoU on PASCAL-Person-Part and CIHP, 75% mIoU on Helen with only 1.95, 3.02, 2.84 and 1.09 clicks per class respectively. These results demonstrate that we can simply acquire high-quality human parsing masks with only a few human effort. We hope this work can motivate more researchers to develop data-efficient solutions to IHP in the future.

In this paper, we aim to devise a new framework to compel the network to be equipped with the capability of detecting objects using image-level class labels as supervision. The challenge of such a weakly supervised setting mainly lies in how to make the network accurately understand both semantics and objectness of a given proposal without bounding box annotations. To this end, we contribute a concise framework, named Class Prototypical Network (CPNet). Concretely, our CPNet defines a set of learnable class prototypes to help classify object proposals. To endow the prototypes be not only discriminative for classes but also sensitive for proposals' objectness, we conduct both class-aware cross-attention and location-aware cross-attention between the feature embeddings of the learnable prototypes and the proposals. The learned attention scores are then used to form the proposal-level category information into the image-level one, making the entire framework be trained without any bounding box annotations. Besides, by applying these two kinds of attention mechanisms, the knowledge from both proposals' location and its class information can be successfully transferred into the corresponding prototypes. With the help of prototypes, our CPNet detects true positive object proposals. In addition, the CPNet further introduces a multi-head detection head to perform complementary training, preventing the model from falling into local discriminative parts and improving the model's performance on challenging non-rigid categories. We examine our CPNet on popular benchmarks,i.e., PASCAL VOC 2007, 2012 and MS COCO 2014. Extensive experiments show our CPNet is a simple and effective framework.

The visible-infrared person re-identification (VI-ReID) is a challenging ReID task, which aims to retrieve and match the same identity's images between the heterogeneous visible and infrared modalities. Thus, the core of this task is to bridge the huge gap between these two modalities. The existing methods mainly face the problem of insufficient perception of modality information, and can not learn good discriminative modality-invariant embeddings for identities, which limits their performance. To solve these problems, we propose a new cross-modality transformer-based method (CMTR) for this visible-infrared person re-identification task, which can explicitly mine the information of each modality and generate better discriminative features based on it. Specifically, to capture inherent characteristics of modalities, we design the novel modality embeddings, which are fused with token embeddings to encode modality information directly. Moreover, to enhance representation of modality embeddings and adjust the distribution of embeddings, we further propose a modality-aware enhancement loss based on the learned modality information, which contains two components to reduce intra-class distance and enlarging inter-class distance simultaneously. To our knowledge, this is the first exploration of applying pure transformer network to the cross-modality re-identification task. We implement extensive experiments on the public SYSU-MM01 and RegDB datasets, and compared with previous methods, our method achieves good performance with more compact and powerful embeddings for the cross-modality retrieval.

Federated recommender system (FRS), which enables many local devices to train a shared model jointly without transmitting local raw data, has become a prevalent recommendation paradigm with privacy-preserving advantages. However, previous work on FRS performs similarity search via inner product in continuous embedding space, which causes an efficiency bottleneck when the scale of items is extremely large. We argue that such a scheme in federated settings ignores the limited capacities in resource-constrained user devices ( i.e. , storage space, computational overhead, and communication bandwidth), and makes it harder to be deployed in large-scale recommender systems. Besides, it has been shown that transmitting local gradients in real-valued form between server and clients may leak users’ private information. To this end, we propose a lightweight federated recommendation framework with privacy-preserving matrix factorization, LightFR , that is able to generate high-quality binary codes by exploiting learning to hash technique under federated settings, and thus enjoys both fast online inference and economic memory consumption. Moreover, we devise an efficient federated discrete optimization algorithm to collaboratively train model parameters between the server and clients, which can effectively prevent real-valued gradient attacks from malicious parties. Through extensive experiments on four real-world datasets, we show that our LightFR model outperforms several state-of-the-art FRS methods in terms of recommendation accuracy, inference efficiency and data privacy.

Metaverse and intelligent transportation system (ITS) are disruptive technologies that have the potential to transform the current transportation system by decreasing traffic accidents and improving driving safety. The integration of Metaverse and transportation technology, called metaverse transportation system (MTS), can greatly improve the intelligence of real transportation system. The digital models built in MTS help to simulate the full life cycle of physical entities, which equip the virtual space with controllability and flexibility. In this article, we concentrate on the field of environment perception, which is the basic function of intelligent vehicles in MTS. To overcome the poor scalability of traditional environment perception methods, we develop the framework of parallel vision for ITS in metaverse (PVITS), consisting of construction of virtual transportation space, model learning based on computational experiments, and feedback optimization based on parallel execution. This article highlights opportunities brought by PVITS in terms of model precision and generalization improvement. Then, the challenges of PVITS are discussed, i.e., distribution difference between virtual and real transportation space, structure design and theoretical interpretation of vision models, and data security and privacy in virtual transportation space. After that, we present several solutions to tackle the application challenges and fully exploit the superior characteristics of PVITS while attenuating their negative side effects. Some potential applications are also given to represent the effectiveness and reliability of PVITS.

Image semantic segmentation aims at the pixel-level classification of images, which has requirements for both accuracy and speed in practical application. Existing semantic segmentation methods mainly rely on the high-resolution input to achieve high accuracy and do not meet the requirements of inference time. Although some methods focus on high-speed scene parsing with lightweight architectures, they can not fully mine semantic features under low computation with relatively low performance. To realize the real-time and high-precision segmentation, we propose a new method named Boundary Corrected Multi-scale Fusion Network, which uses the designed Low-resolution Multi-scale Fusion Module to extract semantic information. Moreover, to deal with boundary errors caused by low-resolution feature map fusion, we further design an additional Boundary Corrected Loss to constrain overly smooth features. Extensive experiments show that our method achieves a state-of-the-art balance of accuracy and speed for the real-time semantic segmentation.

Unsupervised person re-identification (ReID) is a challenging task without data annotation to guide discriminative learning. Existing methods attempt to solve this problem by clustering extracted embeddings to generate pseudo labels. However, most methods ignore the intra-class gap caused by camera style variance, and some methods are relatively complex and indirect although they try to solve the negative impact of the camera style on feature distribution. To solve this problem, we propose a camera-aware style separation and contrastive learning method (CA-UReID), which directly separates camera styles in the feature space with the designed camera-aware attention module. It can explicitly divide the learnable feature into camera-specific and camera-agnostic parts, reducing the influence of different cameras. Moreover, to further narrow the gap across cameras, we design a camera-aware contrastive center loss to learn more discriminative embeddings for each identity. Extensive experiments demonstrate the superiority of our method over the state-of-the-art methods on the unsupervised person ReID task.

The visible-infrared person re-identification (VI-ReID) task aims to retrieve images of pedestrians across cameras with different modalities. In this task, the major challenges arise from two aspects: intra-class variations among images of the same identity, and cross-modality discrepancies between visible and infrared images. Existing methods mainly focus on the latter, attempting to alleviate the impact of modality discrepancy, which ignore the former issue of identity variations and achieve limited discrimination. To address both aspects, we propose a Keypoint-guided Modality-invariant Discriminative Learning (KMDL) method, which can simultaneously adapt to intra-ID variations and bridge the cross-modality gap. By introducing human keypoints, our method makes further exploration in the image space, feature space and loss constraints to solve the above issues. Specifically, considering the modality discrepancy in original images, we first design a Hue Jitter Augmentation (HJA) strategy, introducing the hue disturbance to alleviate color dependence in the input stage. To obtain discriminative fine-grained representation for retrieval, we design the Global-Keypoint Graph Module (GKGM) in feature space, which can directly extract keypoint-aligned features and mine relationships within global and keypoint embeddings. Based on these semantic local embeddings, we further propose the Keypoint-Aware Center (KAC) loss that can effectively adjust the feature distribution under the supervision of ID and keypoint to learn discriminative representation for the matching. Extensive experiments on SYSU-MM01 and RegDB datasets demonstrate the effectiveness of our KMDL method.

Collecting action recognition datasets is time-consuming and labor-intensive. To solve this problem, a few-shot action recognition task that uses episode training to learn the model appears. However, due to the randomness of few-shot learning task sampling, there are great differences between each task, and the characteristics of classes are also diverse. Most of the current methods simply use the same processing flow for action recognition, ignoring the correlation between tasks. To solve this problem, we propose a task adaptive network for few-shot action recognition, which utilizes the dependency of support set and query set categories. Our method mainly includes two key points: Firstly, we add an attention module after the feature extraction module, which can use the attention mechanism to focus the obtained feature representation on more important local information. Secondly, we design a task adaptive module, which uses the support set samples to strengthen all samples of the current task. The module strengthens the common features within each class of the support set and expands the query set to highlight the differences between classes. We have conducted a large number of experiments on two commonly used action recognition data sets: HMDB51 and UCF101. The results of experiments show that our method has strong competitiveness and performs well in the field of few-shot action recognition.

Self-supervised colocalization is to localize common objects in the data set containing only one superclass without using human-annotated labels. Existing methods achieve impressive results by employing self-supervised pretext learning. However, a common limitation still exists. They either tend to overextend activations to the background, or they tend to activate the most discriminative object part. To alleviate this problem, we propose an object representation enhancement model to weaken background distraction and to mine complementary object regions during the object representation learning. Specifically, we first propose an Object-aware Representation Enhancement (ORE) module to estimate an object mask for each input image, guiding the model to disregard the background content and focus on the foreground object. The ORE module and the subsequent self-supervised learning can mutually reinforce each other. Then we propose a Masked Self-supervised Learning branch and design a masked attention consistency objective to induce the model to activate complementary parts of the object effectively. Extensive experiments on four fine-grained data sets demonstrate the superiority of the proposed model.

Cross-view vehicle Reidentification (ReID) has attracted widespread attention as an increasingly important vision task in intelligent transportation and urban surveillance. Benefiting from Convolutional Neural Network (CNN), recent studies have promoted the development of vehicle ReID by extracting discriminative local features. However, two fundamental challenges of small interclass discrepancy caused by different views and large intraclass distance caused by similar appearance still hinder the performance of cross-view vehicle ReID. In this paper, a novel View-aware Sphere Learning Network (VSLN) is proposed to alleviate the above issues while maintaining the merits of CNN-based approaches to generate view-aware sphere-based features. First, a Sphere Feature Embedding Network (SFEN) is proposed to constrain the images into hypersphere for extracting sphere features. On the other hand, this study presents a sphere similarity triple loss to help SFEN concentrate more on robust and discriminative vehicle parts. Second, since the vehicle images are usually captured from different viewpoints, this study further extends SFEN by introducing a Vehicle Viewpoint Predictor (VVP) combined with global attention mechanism to enlarge the discrepancy of interclass and shorten the distance of intraclass. Moreover, a city-scale data set, named Vehicle from Different Viewpoints, containing image-level viewpoint labels, is collected for training VVP. As a result, the proposed VLSN can achieve 96.31% Top-1 accuracy and 79.46% Top-1 accuracy on VeRi-776 and VRIC data sets, respectively. Overall, extensive experimental results on two benchmark data sets show that the proposed VSLN outperforms state-of-the-art methods.

Malware detection is a vital task for cybersecurity. For malware dynamic behavior, threats come from a small number of Application Programming Interfaces (APIs) embedded in the API sequences, which are easily ignored or obfuscated in the detection process. Prior works proposed graph-based learning methods to solve this problem using API-level behavior relations. However, the malware detection is still challenging, due to the ignore of the temporal correlation between malicious behaviors. In this study, we model the software behaviors with multiscaled API graph sequences to represent API-level behaviors as well as graph-level temporal behavior correlations. We then propose a novel Dynamic Evolving Graph Convolutional Network (DEGCN) model to capture dynamic evolving pattern of both local API-level and global graph-level software behaviors. In particular, we first extract the API-level (node) representations to capture the directed graph representations for each time slot. We then propose a Graph-encoding-based Gate Recurrent Unit (GGRU) network to capture the graph-level evolving features and their evolving status. The graph features of different time slots and different graph scales are concatenated to detect whether the software is benign or malicious. Our evaluation with two public benchmarks reports that DEGCN achieves the best performance compared with state-of-the-art algorithms.

The explosion of Internet of Things (IoT) devices will generate massive amounts of data. Due to the limited resources of IoT devices, they usually upload the collected data to edge/cloud servers for processing. To reduce the amount of data uploaded to the edge/cloud server, extracting the features of data on IoT devices has attracted increasing attention. However, most existing related works suffer from two major limitations: 1) difficulty meeting user needs: the extractors they generate are difficult to extract effective features from data on IoT devices and 2) consume a lot of storage resources: they deploy multiple extractors to adapt to the dynamically changing resources of IoT devices. To this end, we propose a nonredundant discriminative feature extraction (DFE) framework, which consists of similarity-based DFE (SDFE) and 2RNestE algorithms. SDFE is first proposed to generate an extractor E that can extract effective discriminative features by rationally exploring the structural information of the data set on the edge/cloud server. Then, 2RNestE is proposed, which takes E as input, and outputs a nonredundant multifunctional extractor by removing redundant subextractors and nesting the remaining nonredundant subextractors together. Finally, the edge/cloud server sends the generated extractor to the IoT device. Experimental results show that the proposed framework reduces memory footprint by about 82.6% and switching overhead by about 84.6% compared with state-of-the-art works.

The high-speed train control system is essential to the safety and efficiency of train operation. With the rapid increase of high-speed railway (HSR) operating mileage and development of information technology, the disposal flow and methods in emergency response are still based on dispatchers and drivers’ experience within the “layered” architecture of current system. There is a certain gap between current processing methods and effective resolution, which may even cause the spread of delay along with the railway networks. Therefore, we propose an integration system of operation control and online rescheduling to improve the recovery ability of HSR carrying capacity. We first describe the framework, information flow, and disposal process of the current system and analyze the shortcomings in handling emergencies. Then, the basic concept, system structure, and framework of the integration system are introduced. Finally, taking temporary speed restriction caused by strong wind as an example, we also analyze the principle of why and how the integration system can promote the recovery ability of HSR carrying capacity.

The coordinated control problem for a multiple high-speed train (HST) system subject to unknown communication delays is systematically investigated in this paper. Taking into consideration the inertial lag of the servo motor, a third-order nonlinear control model is constructed to capture the dynamics of a train in real-world operations. By virtue of the backstepping linearization technique, the coordinated control of trains is formulated as a stabilization problem for a linear multiple-input multiple-output system with an unknown input delay. Distributed control laws with a time-varying low gain parameter are designed, besides solving the stabilization problem, to guarantee a fast convergency rate during the train status adjustment process. Numerical examples are provided to illustrate that the time-varying low gain parameter design achieves better control performance compared with the traditional constant low gain feedback design in terms of the convergency rate and the system overshot, and that the proposed control method is effective in train tracking distance adjustment.

The key to 3D object detection is proper utilization of depth data. Compared with LiDAR based approaches, 3D object detection from a single image remains a challenging task due to the lack of structure information. Recent methods leverage monocular depth estimation as a way to produce 2D depth maps, and adopt the depth maps as additional source of input to explore structure information. However, these methods either encode local structure correlations, or encode long range structure correlations by iteratively passing local messages. In this work, we propose a cross modal attention network (CMAN) for monocular 3D object detection. It is built upon the self-attention module which learns attention map from single modal data. Our CMAN is able to encode structure correlations from depth data, and embed the structure correlations with appearance information which is learned from RGB data. Thanks to the attention learning mechanism, our CMAN learns global structure correlations without iteration. In order to reduce the computational burden, our CMAN adopts a novel node sampler to eliminate redundant nodes during the attention map calculation. Experiment results on benchmark KITTI3D dataset show that our proposed CMAN outperforms the state-of-the-art methods.

With the increasing prevalence of intelligent traffic control and monitoring, research on vehicle re-identification (Re-ID) draws substantial attention in recent years. Different from other cross-view searching tasks such as person Re-ID, the vehicle Re-ID problem is more challenging and unpredictable as viewpoint variations can greatly affect the appearance of vehicles. Existing studies mainly focus on extracting global features based on visual appearance to represent the identity of the target vehicle, while the impact of viewpoint variation is rarely considered. In this paper, we take the view information into account to boost vehicle Re-ID, and introduce latent view labels by clustering and incorporates view information into deep metric learning to tackle the challenge. We also develop a stricter center constraint to further improve the intra-class compactness of feature space. Moreover, we adopt an orthogonal regularization to increase the separability between different vehicles. VARID achieves 79.3% mAP on VeRi-776 and 88.5% mAP on VehicleID which surpasses state-of-the-arts a lot. More comprehensive experimental analyses and evaluations on four benchmarks demonstrate that the proposed method outperforms significantly state-of-the-arts methods.

Unsupervised person re-identification (re-ID) is becoming increasingly popular due to its power in real-world systems such as public security and intelligent transportation systems. However, the person re-ID task is challenged by the problems of data distribution discrepancy across cameras and lack of label information. In this paper, we propose a coarse-to-fine heterogeneous graph alignment (HGA) method to find cross-camera person matches by characterizing the unlabeled data as a heterogeneous graph for each camera. In the coarse-alignment stage, we assign a projection for each camera and utilize an adversarial learning based method to align coarse-grained node groups from different cameras into a shared space, which consequently alleviates the distribution discrepancy between cameras. In the fine-alignment stage, we exploit potential fine-grained node groups in the shared space and introduce conservative alignment loss functions to constrain the graph aligning process, resulting in reliable pseudo labels as learning guidance. The proposed domain adaptation framework not only improves model generalization on target domain, but also facilitates mining and integrating the potential discriminative information across different cameras. Extensive experiments on benchmark datasets demonstrate that the proposed approach outperforms the state-of-the-arts.

Depth estimation is a structure learning problem. The affinity among neighbouring pixels plays an important role in inferring depth values. In this paper, we propose to learn structure affinity in both spatial and temporal domain for accurate depth estimation from monocular videos. Specifically, we first propose a convolutional spatial temporal propagation network (CSTPN) that learns affinity among neighbouring video frames. Secondly, we employ a structure knowledge distillation scheme that transfers the spatial temporal affinity learned by cumbersome network to compact network. By calculating pixel-wise similarities between neighboring frames and neighbouring sequences, our knowledge distillation scheme efficiently captures both short-term and long-term spatial temporal affinity. Finally, we apply a warping loss based on optical flow between video frames to further enforce the temporal affinity. Experiment results show that our proposed depth estimation approach outperform the state-of-the-art methods on both indoor and outdoor benchmark datasets.

The RGB-infrared cross-modality person re-identification (ReID) task aims to recognize the images of the same identity between the visible modality and the infrared modality. Existing methods mainly use a two-stream architecture to eliminate the discrepancy between the two modalities in the final common feature space, which ignore the single space of each modality in the shallow layers. To solve it, in this paper, we present a novel multi-feature space joint optimization (MSO) network, which can learn modality-sharable features in both the single-modality space and the common space. Firstly, based on the observation that edge information is modality-invariant, we propose an edge features enhancement module to enhance the modality-sharable features in each single-modality space. Specifically, we design a perceptual edge features (PEF) loss after the edge fusion strategy analysis. According to our knowledge, this is the first work that proposes explicit optimization in the single-modality feature space on cross-modality ReID task. Moreover, to increase the difference between cross-modality distance and class distance, we introduce a novel cross-modality contrastive-center (CMCC) loss into the modality-joint constraints in the common feature space. The PEF loss and CMCC loss jointly optimize the model in an end-to-end manner, which markedly improves the network's performance. Extensive experiments demonstrate that the proposed model significantly outperforms state-of-the-art methods on both the SYSU-MM01 and RegDB datasets.

Recently, federated learning has been used by Network Intrusion Detection Systems (NIDSs) to expanding data features while preserving data privacy. However, non-independent and identically distributed (non-iid) datasets weaken the model performance of federated learning. In this paper, we present a novel Federated Intrusion Detection System(FIDS) to classify the differences of DDoS attacks from the non-iid dataset. A prototypical weight is introduced to measure the correlations between global data space and local data spaces. We then explore the feature combinations of abnormal behaviors and extract extra features from original data in preprocessing steps. Experimental results show that the FIDS improves the performance of training stability and convergence rate compared to two baselines.

Over the past decades, researches on facial expression recognition have been restricted within six basic expressions (anger, fear, disgust, happiness, sadness and surprise). However, these six words can not fully describe the richness and diversity of human beings' emotions. To enhance the recognitive capabilities for computers, in this paper, we focus on fine-grained facial expression recognition in the wild and build a brand new benchmark FG-Emotions to push the research frontiers on this topic, which extends the original six classes to more elaborate thirty-three classes. Our FG-Emotions contains 10,371 images and 1,491 video clips annotated with corresponding fine-grained facial expression categories and landmarks. FG-Emotions also provides several features (e.g., LBP features and dense trajectories features) to facilitate related research. Moreover, on top of FG-Emotions, we propose a new end-to-end Multi-Scale Action Unit (AU)-based Network (MSAU-Net) for facial expression recognition with image which learns a more powerful facial representation by directly focusing on locating facial action units and utilizing “zoom in” operation to aggregate distinctive local features. As for recognition with video, we further extend the MSAU-Net to a two-stream model (TMSAU-Net) by adding a module with attention mechanism and a temporal stream branch to jointly learn spatial and temporal features. (T)MSAU-Net consistently outperforms existing state-of-the-art solutions on our FG-Emotions and several other datasets, and serves as a strong baseline to drive the future research towards fine-grained facial expression recognition in the wild.

Feature pyramid network (FPN) based models, which fuse the semantics and salient details in a progressive manner, have been proven highly effective in salient object detection. However, it is observed that these models often generate saliency maps with incomplete object structures or unclear object boundaries, due to the indirect information propagation among distant layers that makes such fusion structure less effective. In this work, we propose a novel Cross-layer Feature Pyramid Network (CFPN), in which direct cross-layer communication is enabled to improve the progressive fusion in salient object detection. Specifically, the proposed network first aggregates multi-scale features from different layers into feature maps that have access to both the high- and low- level information. Then, it distributes the aggregated features to all the involved layers to gain access to richer context. In this way, the distributed features per layer own both semantics and salient details from all other layers simultaneously, and suffer reduced loss of important information during the progressive feature fusion. At last, CFPN fuses the distributed features of each layer stage-by-stage. This way, the high-level features that contain context useful for locating complete objects are preserved until the final output layer, and the low-level features that contain spatial structure details are embedded into each layer to preserve spatial structural details. Extensive experimental results over six widely used salient object detection benchmarks and with three popular backbones clearly demonstrate that CFPN can accurately locate fairly complete salient regions and effectively segment the object boundaries.

Multi-view vehicle re-identification (Re-ID) aims to retrieve all images of a target vehicle from a large gallery where the vehicles are captured from non-overlapping cameras. However, the drastic variation in vehicle appearance under different viewpoints greatly affects the performance of the multi-view vehicle Re-ID model, so the key issue in multi-view vehicle Re-ID is learning an effective feature representation that is robust to both dramatic intra-class variability and small inter-class variability. To achieve this goal, we have proposed a multi-center metric learning framework for multi-view vehicle Re-ID. In our approach, we model latent views from vehicle visual appearance directly without any extra labels except ID. Firstly, we introduce several latent view clusters for a vehicle to model latent multi-view information and each view cluster has a learnable center. Then multi-view vehicle matching task can be transformed into two subproblems, cross-view matching and cross-target matching. Finally, an intra-class ranking loss with cross-view center constraint and a cross-class ranking loss with cross-vehicle center constraint are proposed to address the two subproblems, respectively. Extensive experimental evaluations on three widely used benchmarks show the superiority of the proposed framework in contrast to a series of existing state-of-the-arts.

Safe and reliable intelligent charging stations are imperative in an intelligent transportation infrastructure. Over the past few years, a big number of smart charging stations have been deployed, and most of them are online and connected, resulting in potential risks of threats. Although there exists related work on securing intelligent vehicles, very little work focused on the security of charging devices. Unlike traditional network systems, these power-related Industrial Control Systems (ICSs) use many different proprietary protocols and diverse interactions. Traditional anomaly detection methods based on network traffic are thus not suitable for these systems. In this work, we propose an anomaly detection method in real vehicle power supply systems based on a deep architecture model. In particular, we propose a novel traffic anomaly detection model based on Multi-Head Attentions (MHA) that take into account the inherent correlations of traffic generated by ICSs. The MHA model is employed to substitute the traditional feature extraction and rule making process with an acceptable computational cost for classifying traffic data. It is an attention-based model that employs Google Transformer encoder architecture to extract recessive features of traffic for anomaly detection. The effectiveness of the model is demonstrated by experiments on two real-world power ICS testbeds including a substation with a slave charging station and a power generation simulation platform based on a distributed control system. Comprehensive experimental results indicate that the MHA model outperforms the Convolutional Neural Networks (CNN)-based and classical machine learning detection models with an accuracy rate of 99.86%.

Partial Label Learning (PLL) aims to learn from the data where each training example is associated with a set of candidate labels, among which only one is correct. The key to deal with such problem is to disambiguate the candidate label sets and obtain the correct assignments between instances and their candidate labels. In this paper, we interpret such assignments as instance-to-label matchings, and reformulate the task of PLL as a matching selection problem. To model such problem, we propose a novel Graph Matching based Partial Label Learning (GM-PLL) framework, where Graph Matching (GM) scheme is incorporated owing to its excellent capability of exploiting the instance and label relationship. Meanwhile, since conventional one-to-one GM algorithm does not satisfy the constraint of PLL problem that multiple instances may correspond to the same label, we extend a traditional one-to-one probabilistic matching algorithm to the many-to-one constraint, and make the proposed framework accommodate to the PLL problem. Moreover, we also propose a relaxed matching prediction model, which can improve the prediction accuracy via GM strategy. Extensive experiments on both artificial and real-world data sets demonstrate that the proposed method can achieve superior or comparable performance against the state-of-the-art methods.

In blockchain networks, reaching effective supervision while maintaining anonymity to the public has been an ongoing challenge. In existing solutions, certification authorities need to record all pairs of identities and pseudonyms, which is demanding and costly. This article proposed an account guarantee scheme to realize feasible supervision for existing anonymous blockchain networks with lower storage costs. Users are able to guarantee anonymous accounts with account guarantee key pairs generated from certificated polynomial functions, which inherently maintains one-to-n mapping certifications. Single or limited account guarantee key pairs do not leak privacy. Victims are able to request TCs to screen a cheater or disclose a cheater with enough fraud transactions by themselves. Detailed security and privacy analysis showed that the account guarantee scheme preserves user privacy and realizes feasible supervision. Experimental results demonstrated that the account guarantee scheme is efficient and practical.

Current face spoof detection schemes mainly rely on physiological cues such as eye blinking, mouth movements, and micro-expression changes, or textural attributes of the face images [9]. But none of these methods represent a viable mechanism for makeup-induced spoofing, especially since makeup has been widely used. Compared with face alteration techniques such as plastic surgery, makeup is non-permanent and cost efficient, which makes makeup-induced spoofing become a realistic threat to the integrity of a face recognition system. To solve this problem, we propose a generative model to construct spoofing face images (confusing face images) for improving the accuracy and robustness of automatic face recognition. Our network structure is composed of two separate parts, with one using inter-attention mechanism to obtain interested face region, and another using intra-attention to translate imitation style with preserving imitation style-excluding details. These two attention mechanisms can precisely learn imitation style, where inter-attention pays more attention to imitation regions of image and intra-attention learns face attributes with long distance in image. To effectively discriminate generated images, we introduce an imitation style discriminator. Our model (SPGAN) generates face images that transfer the imitation style from target to subject image and preserve the imitation-excluding features. Experimental results demonstrate the performance of our model in improving quality of imitated face images.

Human parsing is an important task in human-centric image understanding in computer vision and multimedia systems. However, most existing works on human parsing mainly tackle the single-person scenario, which deviates from real-world applications where multiple persons are present simultaneously with interaction and occlusion. To address such a challenging multi-human parsing problem, we introduce a novel multi-human parsing model named MH-Parser, which uses a graph-based generative adversarial model to address the challenges of close-person interaction and occlusion in multi-human parsing. To validate the effectiveness of the new model, we collect a new dataset named Multi-Human Parsing (MHP), which contains multiple persons with intensive person interaction and entanglement. Experiments on the new MHP dataset and existing datasets demonstrate that the proposed method is effective in addressing the multi-human parsing problem compared with existing solutions in the literature.

Person re-identification (Re-ID) across multiple datasets is a challenging task due to two main reasons: the presence of large cross-dataset distinctions and the absence of annotated target instances. To address these two issues, this paper proposes a domain adaptive attention learning approach to reliably transfer discriminative representation from the labeled source domain to the unlabeled target domain. In this approach, a domain adaptive attention model is learned to separate the feature map into domain-shared part and domain-specific part. In this manner, the domain-shared part is used to capture transferable cues that can compensate cross-dataset distinctions and give positive contributions to the target task, while the domain-specific part aims to model the noisy information to avoid the negative transfer caused by domain diversity. A soft label loss is further employed to take full use of unlabeled target data by estimating pseudo labels. Extensive experiments on the Market-1501, DukeMTMC-reID and MSMT17 benchmarks demonstrate the proposed approach outperforms the state-of-the-arts.

Learning-based approaches to graph matching have been developed and explored for more than a decade, have grown rapidly in scope and popularity in recent years. However, previous learning-based algorithms, with or without deep learning strategy, mainly focus on the learning of node and/or edge affinities generation, and pay less attention on the learning of the combinatorial solver. In this paper we propose a fully trainable framework for graph matching, in which learning of affinities and solving for combinatorial optimization are not explicitly separated as in many previous arts. We firstly convert the problem of building node correspondences between two input graphs to the problem of selecting reliable nodes from a constructed assignment graph. Subsequently, the graph network block module is adopted to perform computation on the graph to form structured representations for each node. It finally predicts a label for each node that is used for node classification, and the training is performed under the supervision of both permutation differences and the one-to-one matching constraints. The proposed method is evaluated on four public benchmarks in comparison with several state-of-the-art algorithms, and the experimental results illustrate its excellent performance.

Partial Multi-Label learning (PML) learns from the ambiguous data where each instance is associated with a candidate label set, where only a part is correct. The key to solve such problem is to disambiguate the candidate label sets and identify the correct assignments between instances and their ground-truth labels. In this paper, we interpret such assignments as instance-to-label matchings, and formulate the task of PML as a matching selection problem. To model such problem, we propose a novel grapH mAtching based partial muLti-label lEarning (HALE) framework, where Graph Matching scheme is incorporated owing to its good performance of exploiting the instance and label relationship. Meanwhile, since conventional one-to-one graph matching algorithm does not satisfy the constraint of PML problem that multiple instances may correspond to multiple labels, we extend the traditional probabilistic graph matching algorithm from one-to-one constraint to many-to-many constraint, and make the proposed framework to accommodate to the PML problem. Moreover, to improve the performance of predictive model, both the minimum error reconstruction and k-nearest-neighbor weight voting scheme are employed to assign more accurate labels for unseen instances. Extensive experiments on various data sets demonstrate the superiority of our proposed method.

Big data for social transportation brings unprecedented opportunities for us to solve the transportation problems that cannot be solved by traditional methods and build the next generation of the intelligent transportation system (ITS). As one of the important functions of the ITS, supply-demand difference prediction for autonomous vehicles provides a decision basis for its control. In this article, a new learning process is proposed with Multiple feature Extraction and Fusion utilizing the combination of deep and shallow Features (MEFF) (the spatial deep features, (short and long) temporal deep features, and fuzzy shallow (semantic) features). The spatial deep features are captured with residual network and dimension reduction in spatial deep block. The fuzzy shallow (semantic) features are captured with multiattention fuzzy mechanism in the fuzzy shallow block. With the fused spatial deep features and fuzzy shallow features, the temporal deep features are captured with long short-term memory (LSTM) and attention mechanism in the temporal and prediction block to get the final prediction results. Based on two different distributions of membership attention (mean distribution and Gaussian distribution) in the fuzzy shallow block, our process MEFF has two methods, i.e., MEFF-mean method and MEFF-Gaussian method. Extensive experiments show that our methods provide more accurate and stable prediction results than the existing state-of-art-methods.

Partial label learning (PLL) aims to learn from the data where each training instance is associated with a set of candidate labels, among which only one is correct. Most existing methods deal with this type of problem by either treating each candidate label equally or identifying the ground-truth label iteratively. In this article, we propose a novel PLL approach named HERA, which simultaneously incorporates the HeterogEneous Loss and the SpaRse and Low-rAnk procedure to estimate the labeling confidence for each instance while training the desired model. Specifically, the heterogeneous loss integrates the strengths of both the pairwise ranking loss and the pointwise reconstruction loss to provide informative label ranking and reconstruction information for label identification, whereas the embedded sparse and low-rank scheme constrains the sparsity of ground-truth label matrix and the low rank of noise label matrix to explore the global label relevance among the whole training data, for improving the learning model. Comprehensive ablation study demonstrates the effectiveness of our employed heterogeneous loss, and extensive experiments on both artificial and real-world datasets demonstrate that our method achieves superior or comparable performance against state-of-the-art methods.

This paper addresses an error-driven nonlinear feedback design technique to improve the dynamic performance of fuzzy adaptive dynamic surface control (DSC) for a class of uncertain multiple-input-multiple-output nonlinear systems with prescribed tracking performance. The highlight of the error-driven nonlinear feedback technique is that the feedback gain self-regulates versus different levels of output and virtual tracking errors, this reflects the classical control design criterions commendably: relatively high feedback gains can be implemented to guarantee disturbances and uncertainties attenuation and so on to improve the control performance when small tracking errors are measured, and relatively small feedback gains can be implemented to circumvent the problems of actuator and states saturations when large tracking errors are measured. The complexity problem of the traditional backstepping design is circumvented owe to the peculiarity of DSC method. Caused by the compound error functions of nonlinear feedback dynamics, a nonquadratic Lyapunov function is used to deduce the conditions of closed-loop stability. Fuzzy logic systems and error transformation-based method are used in the online learning of completely unknown dynamics and the prescribed performance tracking, respectively. Comparative results are presented to demonstrate the effectiveness and preponderance of the proposed control scheme with comparison to existing ones.

Vehicle re-identification (ReID) aims to identify a target vehicle in different cameras with non-overlapping views, and it plays an important role when the car licence plate recognition is unavailable or unreliable. Compared with face recognition and person ReID tasks, it is difficult to train an effective vehicle ReID model due to two reasons: the different views greatly affect the visual appearance of a vehicle, and different vehicles may exhibit fairly similar visual appearance when their images are captured from one unified single view. To handle these training difficulties, we introduce several latent groups to represent multiple views. Then, the vehicle ReID problem is modeled as two sub tasks, including matching vehicles in a same view and across different views. A fine-grain ranking loss and a relative coarse-grain ranking loss are proposed to each task respectively. Extensive experimental analyses and evaluations on two benchmarks demonstrate the proposed method can achieve state-of-the-art performance.

The use of Partial Differential Equation models for studying traffic flows has a fairly long history. This paper deals with the Cauchy problem of a two-component Camassa-Holm equation. First, we prove that the solution ρ keeps the property of having compact support for any further time provided the initial data ρ_0 has compact support. While the initial data m_0 has compact support then the solution m will remain compactly supported, only if ρ is also initially compactly supported. Then, we get the infinite propagation speed in the sense that the solution u with compactly supported initial data does not have compact support any longer in its lifespan. Although the nontrivial solution u is no longer compactly supported, a detailed description about the profile of the solution u is shown as it evolves over time.

As one of the important functions of the intelligent transportation system (ITS), supply-demand prediction for autonomous vehicles provides a decision basis for its control. In this paper, we present two prediction models (i.e. ARLP model and Advanced ARLP model) based on two system environments that only the current day's historical data is available or several days' historical data are available. These two models jointly consider the spatial, temporal, and semantic relations. Spatial dependency is captured with residual network and dimension reduction. Short term temporal dependency is captured with LSTM. Long term temporal dependency and temporal shifting are captured with LSTM and attention mechanism. Semantic dependency is captured with multi-attention mechanism. Extensive experiments show that our frameworks provide more accurate prediction results than the existing methods.

Semantic image synthesis via text description is a desirable and challenging task, which requires more protection of the text irrelevant content in the original image. Existing methods directly modify the original image, which become more difficult when encountering high resolution image, and the generated images are also blurred and lack in detail. This paper presents a novel network architecture to progressively manipulate an image starting from low-resolution, while introducing the original image of corresponding size at different stages with our proposed union module to avoid losing of detail. And the progressive design of the network allows us to modify the image from coarse into fine. Compared with the previous methods, our new method can successfully manipulate a high resolution image and generate a new image with background protection and fine details. The experimental results on CUB-200-2011 dataset show that the proposed approach outperforms existing methods in terms of image detail, background protection and high resolution generation.

Recently, co-saliency detection, which aims to automatically discover common and salient objects appeared in several relevant images, has attracted increased interest in the computer vision community. In this article, we present a novel graph-matching based model for co-saliency detection in image pairs. A solution of graph matching is proposed to integrate the visual appearance, saliency coherence, and spatial structural continuity for detecting co-saliency collaboratively. Since the saliency and the visual similarity have been seamlessly integrated, such a joint inference schema is able to produce more accurate and reliable results. More concretely, the proposed model first computes the intra-saliency for each image by aggregating multiple saliency cues. The common and salient regions across multiple images are thus discovered via a graph matching procedure. Then, a graph reconstruction scheme is proposed to refine the intra-saliency iteratively. Compared to existing co-saliency detection methods that only utilize visual appearance cues, our proposed model can effectively exploit both visual appearance and structure information to better guide co-saliency detection. Extensive experiments on several challenging image pair databases demonstrate that our model outperforms state-of-the-art baselines significantly.

Tracking planar objects has a wide range of applications in robotics. Conventional template tracking algorithms, however, often fail to observe fast object motion or drift significantly after a period of time, due to drastic object appearance change. To address such challenges, we propose a novel constrained confidence matching algorithm for motion estimation and a robust Kalman filter for template updating. Integrated with an accurate occlusion detector, our approach achieves accurate motion estimation in presence of partial occlusion, by excluding occluded pixels from computation of motion parameters. Furthermore, the proposed Kalman filter employs a novel control-input model to handle the object appearance change, which brings our tracker high robustness against sudden illumination change and heavy motion blur. For evaluation, we compare the proposed tracker with several state-of-the-art planar object trackers on two public benchmark datasets. Experimental results show that our algorithm achieves robust tracking results against various environmental variations, and outperforms baseline algorithms remarkably on both datasets.

Heterogeneous Face Recognition (HFR) refers to the problem of recognizing faces across different visual domains and has attached great attention owing to its tremendous potential benefits in practical applications. In this paper, a novel feature learning approach named hierarchical discriminant feature learning (HDFL) has been proposed for HFR. Different from traditional feature learning based HFR approaches, the proposed HDFL aims to learn the most discriminative information via a two-layer hierarchical boosting network (HBN), where the hierarchical discriminative information can be exploited in the learned features and the appearance difference can be effectively reduced, simultaneously. Extensive experiments on three different heterogeneous face databases demonstrate that our approach consistently outperforms the state-of-the-art methods.

The information exchange gap between current operation control and dynamic scheduling in high-speed railway systems (HRSs) still exists, and this gap has hindered the further integrative improvement of HRSs. This paper aims to explore a feasible solution to bridging the information exchange gap for further improving the efficiency of HRSs, with the parallel intelligent systems for integrated HRS operation control and dynamic scheduling first analyzed and constructed using the ACP approach, that is, "artificial systems" (A), "computational experiments," (C) and "parallel execution" (P). Then, on the basis of the constructed parallel intelligent systems, experiments on several typical scenarios in HRSs are conducted to achieve a set of control and management strategies for actual HRSs. Experimental results show that a number of powerful tools provided by the proposed parallel intelligent systems can be utilized not only to study the current HRSs, but also to further undertake research on integrated operation control and dynamic scheduling for HRSs.

Behavior detection is an important research field in pattern recognition and it can be applied to control the traffic safety. The mobile phones becomes a potential security when the pedestrians unconsciously use the phones during the crossing of the street. To avoid the traffic accidents, this paper proposes a new algorithm of the pedestrian behavior detection for the people unconsciously using phones. Firstly, the method that pedestrian detection based on gradient and texture feature integration is used to find the position of pedestrian. Secondly, Selective search is used to get the position of sensitive parts. We choose the arms as the sensitive parts in this algorithm. Finally, we extract and classify sensitive parts. Currently, fewer people are studying this topic. Therefore, we construct a new pedestrian image set that contains 2.5G images called PWUM (Pedestrian Who use mobile phone) set for verifying the effectiveness of our algorithm. Experimental results show that the proposed algorithm can efficiently detect pedestrian who is using mobile phone on PWUM dataset.

Power demand forecasting plays a very important role in many electricity-required industries, such as modern high-speed railways or urban railways. Accurate forecasting will guarantee that electrical equipments such as electric traction systems for trains work under safe, robust and efficient status. Recently, many studies adopt the learning-based methods to achieve the prediction of power demand. However, most of the studies use the traditional classification or clustering algorithms which may not satisfy the requirements of accuracy and efficiency due to the complex features in smart grid. In this paper, we focus on solving the power demand forecasting problem based on deep learning structures. We first propose a deep learning based framework for power demand forecasting with Deep Belief Network (DBN). Then, we use an algorithm called Adaboost to combine weak learners with strong learners, which can increase the accuracy significantly in real-world scenarios. The prediction of the load status is realized by analyzing the information of historical distribution transformer load, weather, electricity population and some other related information. It is also worth noting that the training process of these DBN networks can be parallel, which effectively shorten the processing time and provide the possible of real-time predicting. Our experiment on real-world data from the electrical company shows results that the deep leaning based methods can increase the accuracy of forecasting and significantly shorten the prediction time.

This paper mainly presented a system which can make a prediction to the distribution transformer's load status in smart grid. Since the operation of distribution transformer's load status is generally in the post processing stage at the current stage, lacking forecasting work on distribution transformer's operation and load status. Given the issues above, to reduce costs, ensure the security of power supply, and improve the emergency response capabilities, we presented a prediction system, which can predict the load status of distribution transformer by utilising the data mining algorithm. Besides, the system also provides a platform for the management and maintenance of electrified wire netting's information. In this system, users can conveniently manage the vast and multifarious data sets.

Microarrays are one of the latest breakthroughs in experimental molecular biology, which already provide huge amount of valuable gene expression data. Biclustering algorithm was introduced to capture the coherence of a subset of genes and a subset of conditions. In this paper, we presented a MIWB algorithm to find biclusters of gene expression data. MIWB algorithm uses the weighted mutual information as similarity measure which can be simultaneously detected complex linear and nonlinear relationships between genes. Our algorithm first used the weighted mutual information to construct the seed gene set of each biculster, then we calculated each gene's probability belonging to each bicluster and complete the initial partition of genes set utilizing the given threshold, then by optimising the objective function we completed weights update and conditions set selection, by further repartition of the entire dataset and optimization of biclusters we obtained the final biclusters. We evaluated our algorithm on yeast gene expression dataset, and experimental results show that MIWB algorithm can generate large capacity biclusters with lower mean squared residue.

Combinatorial Map (CM) is becoming increasingly popular due to its power in modeling topological structures with subdivided objects, which is widely used in the fields of social network, computer vision, social media and so on. However, due to its specific structural properties, an unprotected release of a combinatorial map may cause the identity disclosure problem, which is a major privacy breach revealing the identification of entities with certain background knowledge known by an adversary. In this paper, we discuss the privacy preserving problem in publishing private combinatorial maps. We first formalize a specific anonymizing model to deal with dart-related attacks, and discuss an efficient metric to quantify information loss incurred in the perturbation. Then we propose an efficient method for the dart anonymization problem to prevent a CM from the attack. Our approaches are efficient and practical, and have been validated by extensive experiments on two sets of synthetic data.

Recently, the development of cloud computing has received considerable attention. For cloud service providers, packing VMs onto a small number of servers is an effective way to reduce energy costs, so as to improve the efficiency of the data center. However allocating too many VMs on a physical machine may cause some hot spots which violate the SLA of applications. Load balancing of the entire system is hence needed to guarantee the SLA. In this paper, we present a simulated-annealing load balancing algorithm for solving the resource allocation and scheduling problem in a cloud computing environment. Experimental results show that this method is able to achieve load balancing, and performs better than the round robin and basic simulated-annealing algorithms.

Cloud computing is becoming increasingly popular due to its power in providing high-performance and flexible service capabilities. More and more internet users have accepted this innovative service model and been using various cloud-based services every day. However, these service-using data is quite valuable for marketing purposes, as it can reflect a user's interest and service-using pattern. Therefore, the privacy issues have been brought out. Recently, many studies focus on access control and other traditional security problems in cloud, and little studied on the topic of the private service data publishing. In this paper, we study the private service data publishing problem by representing the data with a hyper graph, which is quite efficient to illustrate complex relationships among users. We first formulate the problem with a popular background knowledge attack model named rank attack, and then provide an anonymization-based method to prevent the released data from such attacks. We also take data utility into consideration by defining specific information loss metrics. The performances of the methods have been validated by two sets of synthetic data.

Railway dangerous goods tracing is a typical application of the sensor network. Application correlation among the sensor, carriage and train is a graph relationship which can be described by using RDF frameworks. It requires data management system to manage a large scale of ever-increasing RDF data, and support semantic access for monitoring the safety state of the environment inside the carriage. For these problems, this paper proposes RDF data query and management method based on HBase and structure index, and optimization method of query engine. The method is enforced by rewriting SPARQL statements according of correlation degree between them, and at querying time, "structure-level" index is used to identify the groups of RDF data, then the "data-level" data matching utilizes the proposed scalable storage mechanism based on hash-oriented multiple table partition of data entity class. As shown in our experiments, our approach can effectively reduce the semantic query time, enhance storage scalability and effective support multi-criteria query of sensor data.

Data publishing based on hypergraphs is becoming increasingly popular due to its power in representing multirelations among objects. However, security issues have been little studied on this subject, while most recent work only focuses on the protection of relational data or graphs. As a major privacy breach, identity disclosure reveals the identification of entities with certain background knowledge known by an adversary. In this paper, we first introduce a novel background knowledge attack model based on the property of hyperedge ranks, and formalize the rank-based hypergraph anonymization problem. We then propose a complete solution in a two-step framework: rank anonymization and hypergraph reconstruction. We also take hypergraph clustering (known as community detection) as data utility into consideration, and discuss two metrics to quantify information loss incurred in the perturbation. Our approaches are effective in terms of efficacy, privacy, and utility. The algorithms run in near-quadratic time on hypergraph size, and protect data from rank attacks with almost the same utility preserved. The performances of the methods have been validated by extensive experiments on real-world datasets as well. Our rank-based attack model and algorithms for rank anonymization and hypergraph reconstruction are, to our best knowledge, the first systematic study to privacy preserving for hypergraph-based data publishing.

For urban metro systems with platform screen doors, train automatic stop control (TASC) has recently attracted significant attention from both industry and academia. Existing solutions to TASC are challenged by uncertain stopping errors and the fast decrease in service life of braking systems. In this paper, we try to solve the TASC problem using a new machine learning technique and propose a novel online learning control strategy with the help of the precise location data of balises installed at stations. By modeling and analysis, we find that the learning-based TASC is a challenging problem, having characteristics of small sample sizes and online learning. We then propose three algorithms for TASC by referring to heuristics, gradient descent, and reinforcement learning (RL), which are called heuristic online learning algorithm (HOA), gradient-descent-based online learning algorithm (GOA), and RL-based online learning algorithm (RLA), respectively. We also perform an extensive comparison study on a real-world data set collected in the Beijing subway. Our experimental results show that our approaches control all stopping errors in the range of ±0.30 m under various disturbances. In addition, our approaches can greatly increase the service life of braking systems by only changing the deceleration rate a few times, which is similar to experienced drivers. Among the three algorithms, RLA achieves the best results, and GOA is a little better than HOA. As online learning algorithms can dynamically reduce stopping errors by using the precise location data from balises, it is a promising technique in solving real-world problems.

The multicast problem in multi-channel multi-radio wireless mesh networks has received much attention recently. Most recent studies on this problem focus on improving the network throughput. However, many real-world applications require routing algorithms to achieve low-delay and low-loss. In this paper, we tackle the problem of constructing a robust minimum-cost multicast tree that tolerates link interference. To save bandwidth resource and alleviate the interference in the communication, we propose a robust multicast algorithm for multi-channel multi-radio wireless mesh networks. Our experimental results show that our algorithm is very efficient to achieve better performances in network throughput and end-to-end delay than previous studies.

Focusing on the fact that the grid users may require various QoS and have quite different preferences on each QoS, a User Preference Driven Approach(UPDA) is proposed to schedule grid tasks with multi-QoS constraints. In UPDA, user's needs and preferences are modeled into matrixes to serve as the basis of scheduling and Analytical Hierarchy Process(AHP) method is adopted to perform a pair wise comparison between different kinds of QoS demands to gain the preference on each QoS. Furthermore, UPDA tries to offer optimal performance on the multi-QoS targets with the consideration of the resource utilization, which is just in compliance with the ground rule of fully utilizing of spare resource in grid. Simulation results confirm that UPDA can not only better satisfy the diverse requirements of users, but also perform well in term of the resource utilization rate from the system's perspectives.

Scheduling mechanism in cloud computing environment concerns more about the needs of users due to the commercial purpose of cloud computing. In response to this user-oriented feature, an Adaptive Resource Scheduling Mechanism based on User Behavior Feedback(ARSM-UBF) is developed to schedule cloud resources from the perspective of users. In ARSM-UBF, information collected from scheduling interaction is adopted as the reflection of user behavior and by introducing the relevance feedback mechanism, user behavior information is integrated into the resource scheduling and continuous feedback is proceeded on the following scheduling process to make scheduling results closer to the user's subjective perception. Furthermore, ARSM-UBF gives attention to the efficiency of cloud system. Simulation results confirm that this user behavior feedback based scheduling mechanism is much more efficient since it can not only provide guaranteed services for users with diverse requirements, but also improve the resource utilization rate from the system's perspectives.

Combining manifold ranking with active learning (MRAL for short) is one popular and successful technique for relevance feedback in content-based image retrieval (CBIR). Despite the success, conventional MRAL has two main drawbacks. First, the performance of manifold ranking is very sensitive to the scale parameter used for calculating the Laplacian matrix. Second, conventional MRAL does not take into account the redundancy among examples and thus could select multiple examples that are similar to each other. In this work, a novel MRAL framework is presented to address the drawbacks. Concretely, we first propose a self-tuning manifold ranking algorithm that can adaptively calculate the Laplacian matrix via a local scaling mechanism, and then develop a hybrid active learning algorithm by integrating three well-known selective sampling criteria, which is able to effectively and efficiently identify the most informative and diversified examples for the user to label. Experiments on 10,000 Corel images show that the proposed method is significantly more effective than some existing approaches.

Many real applications, such as network traffic monitoring, intrusion detection, satellite remote sensing, and electronic business, generate data in the form of a stream arriving continuously at high speed. Clustering is an important data analysis tool for knowledge discovery. Compared with traditional clustering algorithms, clustering stream data is an important and challenging problem which has attracted many researchers. Clustering stream data is facing two main challenges. First, as the data is continuously arriving with high rate and the computer storage capacity is limited, raw data can only be scaned in one pass. Second, stream data is always changing with time, so viewing a data stream as a set of static data can deteriorate the clustering quality. In fact, users are more concerned with the evolving behaviors of clusters which can help people making correct decisions. This paper proposes a density-grid based clustering algorithm, PKS-Stream-I, for stream data. It is an optimization of PKS-Stream in density detection period selection, sporadic grid detection and removal. Empirical results show the proposed method yields out better performance.

Chord is a popular and successful topology for Peer-to-Peer (P2P) data sharing. However, the conventional chord is challenged by two main drawbacks. First, it fails to consider the physical topology of the P2P network for designing the lookup solution, which may bring tremendous delay to network routing. Second, its performance of is usually limited by the high space complexity of data storage and thus data retrieval may suffer further network delay. In this work, we propose an improved chord based on Counting Bloom Filter and topology aware lookup to address the drawbacks. We first apply counting Bloom filter for data storage to reduce the space complexity. We then develop a topology-aware lookup mechanism to further speed up the search for local resources. Simulation results show that our improved chord scheme is significantly more efficient than the conventional chord method.

Data publishing based on hyper graphs is becoming increasingly popular due to its power in representing multi-relations among objects. However, security issues have been little studied on this subject, while most recent work only focuses on the protection of relational data or graphs. As a major privacy breach, identity disclosure reveals the identification of entities with certain background knowledge known by an adversary. In this paper, we first introduce a novel background knowledge attack model based on the property of hyper edge ranks, and formalize the rank-based hyper graph anonymization problem. We then propose a complete solution in a two-step framework, with taking community preservation as the objective data utility. The algorithms run in near-quadratic time on hyper graph size, and protect data from rank attacks with almost same utility preserved. The performances of the methods have been validated by extensive experiments on real-world datasets as well.

As an integral part of data security, identity disclosureis a major privacy breach, which reveals the identification of entities with certain background knowledge known by an adversary. Most recent studies on this problem focus on the protection of relational data or simple graph data (i.e. undirected, un weighted and acyclic). However, a weighted graph can introduce much more unique information than its simple version, which makes the disclosure easier. As more real-world graphs or social networks are released publicly, there is growing concern about privacy breaching for the entities involved. In this paper, we first formalize a general anonymizing model to deal with weight-related attacks, and discuss an efficient metric to quantify information loss incurred in the perturbation. Then we consider a very practical attack based on the sum of adjacent weights for each vertex, which is known as volume in graph theory field. We also propose a complete solution for the weight anonymization problem to prevent a graph from volume attack. Our approaches are efficient and practical, and have been validated by extensive experiments on both synthetic and real-world datasets.

Data Swapping is a popular value-invariant data perturbation technique. The quality of a data swapping method is measured by how well it preserves data privacy and data utility. As swapping data globally is computationally impractical, to guarantee its performance in these metrics appropriate, localization schemes are often conducted in advance. Equi-depth partitioning is preferred by most of the existing data perturbation techniques as it provides uniform privacy protection for each data tuple. However, this method performs ineffectively for two types of applications: one is to maintain statistics based on equi-width partitioning, such as the multivariate histogram with equal bin width, and the other is to preserve parametric statistics, such as covariance, in the context of sparse data with non-uniform distribution. As a natural solution for the above application, this paper explores the possibility of using data swapping with equi-width partitioning for private data publication, which has been little used in data perturbation due to the difficulty of preserving data privacy. With extensive theoretical analysis and experimental results, we show that, Equi-Width Swapping (EWS)can achieve a similar performance in privacy preservation to that of Equi-Depth Swapping (EDS) if the number of partitions is sufficiently large (e. g. ¿ = ¿N, where N is the size of dataset). Our experimental results in both synthetic and real-world data validate our theoretical analysis.