Multi-view learning aims to effectively integrate data from different sources by exploring the consistency and complementarity across views. Current multi-view methods based on Graph Convolutional Networks (GCNs) primarily focus on local information, making it difficult to capture global dependencies. Furthermore, multi-view data typically lack explicit structural representations, and the topologies constructed via node similarity in existing approaches are prone to noise, while simple fusion strategies are often inadequate for effectively suppressing this noise and for uncovering meaningful structural information. To tackle these issues, this paper proposes CoGFormer, a cooperative graph transformer with structural consensus learning. CoGFormer maps multi-view data into a unified space and jointly models local and global consensus: a denoising structural consensus graph convolutional network refines the consensus graph to enhance local consistency and robustness, while a structure-guided attention mechanism explicitly injects high-order cross-view structural biases to capture global consistency and improve semantic coherence. Experiments on multiple benchmarks demonstrate that CoGFormer outperforms existing state-of-the-art methods, validating its effectiveness.

Multimedia technologies leverage multi-source to alleviate real-world data incompleteness, providing a versatile platform for multi-view learning. Among existing research, graph-based multi-view learning has achieved notable success. However, prior studies always immerse in comprehensive collaboration across all views and nodes to pursue consistency and complementary, which ignore the negative contribution of nodes from low-quality views. To overcome the above limitation, we explore node behavior selection in multi-view dynamic modeling and propose a knowledge-aware multi-view state space model. Specifically, nodes autonomously select either activation sequences or static sequences according to their current knowledge. In the former, we design the mask-based attention mechanism to capture the dynamics of node behaviors. In the latter, we construct a history pool and simulate synaptic signals to regulate the behavioral distribution of nodes. Moreover, the proposed model provides a directional inter-view diffusion equation that selectively propagates information to alleviate interference from low-quality nodes across views. Extensive experiments demonstrate that the proposed model outperforms baselines on multiple benchmarks and achieves significant performance improvement.

Multi-view representation learning, which utilizes multiple channels to improve perceptual accuracy, is recognized for its effectiveness in the analysis of multi-view data. However, deploying these methods in real-world scenarios presents two primary challenges. 1) Lack of Variegation: Multi-view representation techniques commonly observe along a singular axis, i.e., the attribute axis; 2) Insufficient Relationship: Most multi-view models lack mechanisms for exploring potential relationships between attribute axis and channel axis. To mitigate these obstacles, we design a Dual Impulse Network framework for multi-view representation learning (DIN) to train a feature representation. In this framework, a strategy observed along the channel axis and attribute axis simultaneously is introduced, and two different representations are generated by two analogous impulse networks, which are capable of extracting information corresponding to different axes. Furthermore, we incorporate an integration network that analyzes the potential relationship between attribute axis and channel axis to generate two attention matrices. The final two feature representations derived from these attention matrices are aggregated to amplify the expression of internal information. Comprehensive experimental results support the efficacy and superiority of the proposed framework, demonstrating improvements in classification performance compared to state-of-the-art methods.

User preference modeling based on hypergraphs has shown significant potential in recommender systems. However, existing methods model complex higher-order relations rely on existing hypergraph structures, such well-constructed hypergraphs are not readily accessible in every situation. Furthermore, since existing methods perform message-passing based on the same hypergraph convolution function, they often overlook diverse relation patterns, thus lacking precision. In this work, we propose an Enhanced Recommendation Framework with Hypergraph Mixture of Experts (HMoRec). Specifically, we first employ a sparse optimal transport clustering mechanism to generate high-quality hypergraph without requiring external knowledge. Then, we model diverse higher-order interactions and enhance representation learning based on the hypergraph mixture of experts and cross-view representation fusion. Extensive experiments on four real-world multi-domain datasets have shown that our HMoRec achieves significant performance gains.

Graphs with abundant attributes are essential in modeling interconnected entities and enhancing predictions across various real-world applications. Traditional Graph Neural Networks (GNNs) often require re-training for different graph tasks and datasets. Although the emergence of Large Language Models (LLMs) has introduced new paradigms in natural language processing, their potential for generic graph mining-training a single model to simultaneously handle diverse tasks and datasets-remains under-explored. To this end, our novel framework ${\sf MuseGraph}$MuseGraph, seamlessly integrates the strengths of GNNs and LLMs into one foundation model for graph mining across tasks and datasets. This framework first features a compact graph description to encapsulate key graph information within language token limitations. Then, we propose a diverse instruction generation mechanism with Chain-of-Thought (CoT)-based instruction packages to distill the reasoning capabilities from advanced LLMs like GPT-4. Finally, we design a graph-aware instruction tuning strategy to facilitate mutual enhancement across multiple tasks and datasets while preventing catastrophic forgetting of LLMs' generative abilities. Our experimental results demonstrate significant improvements in five graph tasks and ten datasets, showcasing the potential of our ${\sf MuseGraph}$MuseGraph in enhancing the accuracy of graph-oriented downstream tasks while improving the generation abilities of LLMs.

Temporal knowledge graph forecasting (TKGF) has become a crucial tool for predicting future facts based on temporal knowledge graphs. Traditional methods face inherent limitations: rule-based reasoning approaches are restricted to exploring events that are related to historical rules and fail to fully utilize global graph information. In contrast, although embedding-based methods have the ability to capture global graph information through vector representations, they struggle to effectively incorporate historical rules. To address these issues, this article proposes a novel framework that combines historical rule-based reasoning with embedding-based prediction enhanced by curriculum learning (CL) to predict unknown events. The framework consists of two complementary modes: the rule mode, which extracts temporal rules through time-filtered weighted sampling and applies them for prediction, and the embedding mode, which progressively represents entities, relations, and time as vectors from simple to complex through CL and generates predictions based on these embedding vectors. The framework effectively integrates rules with the entire graph information by combining the predictions from two models, resulting in a more comprehensive prediction. Extensive experiments on multiple public datasets demonstrate that the proposed method outperforms state-of-the-art approaches in terms of performance, validating the effectiveness of its hybrid approach, and highlighting the significant potential of CL in TKGF.

Unsupervised person re-identification (USL-ReID) typically relies on clustering to generate pseudo-labels, but significant cross-view appearance variations often cause images of the same identity to be split into different clusters. Training on such noisy pseudo-labels severely degrades the learned representations. Therefore, learning robust view-invariant features is paramount. Data augmentation provides a direct way to enhance invariance, yet its trade-offs in USL-ReID remain under-explored: weak augmentations usually preserve identity semantics but lack diversity, whereas strong augmentations provide richer appearance diversity at the cost of partially corrupting identity-consistent semantic cues. To address this challenge, we propose Invariant Representation learning with Progressive Prototype Refinement (IRPP), a unified framework that learns invariant and discriminative features from noisy pseudo-labels. IRPP consists of three synergistic components. First, an Augmented Dual-Contrastive Learning (ADCL) module performs dataset-level prototype-guided invariant learning by contrasting weakly and strongly augmented views against cluster-derived prototypes. Second, an Alignment and Uniformity Learning (AUL) module regularizes the mini-batch-level weak-strong feature geometry, leading to more stable feature distributions under data augmentation. Third, a Progressive Prototype Refinement (PPR) mechanism progressively optimizes cluster centroids into cleaner prototypes, thereby mitigating the influence of noisy pseudo-labels and further strengthening invariant representation learning. This closed-loop design enables prototype-guided contrastive learning, weak-strong regularization, and prototype refinement to mutually reinforce each other. Extensive experiments on standard USL-ReID benchmarks demonstrate that IRPP achieves state-of-the-art performance with a simple and efficient training pipeline. Code is available at https://github.com/Trangle12/IRPP.

Graph Neural Networks (GNNs) have become a powerful tool for learning representations from graph-structured data, leveraging the relationships between nodes and their features. Despite their success, they often lack interpretability due to the black-box nature of neural networks, and further development may be limited. Moreover, previous GNN-based multi-view methods typically rely on simple feature fusion techniques such as weighted averaging or concatenation, which fail to capture the complex dependencies between views. In this paper, we propose a novel framework, namely Interpretable Multi-View Feature Representation via physical partial differential equation (IMvFR), to address these limitations in the context of multi-view semi-supervised learning. By integrating GNNs with partial differential equations (PDEs), we model the evolution of multi-view feature representations as a dynamic process. This provides a natural and interpretable framework for understanding how information flows between different views, overcoming the black-box nature of traditional GNNs. Additionally, we formulate multi-view feature representations as an initial-value problem within the framework of PDEs, providing a clear and interpretable mechanism for label propagation and feature fusion, thus facilitating the acquisition of global and local information between views. Comprehensive experimental results on eight datasets demonstrate that the proposed method achieves superior performance compared with state-of-the-art methods.

Data missing is a common issue in real-world applications, posing significant challenges for incomplete data processing. Traditional incomplete multi-view clustering methods rely on manually-designed optimization problems based on prior interpretable knowledge, considering the full utilization of available data. However, their limited feature extraction capability may become a bottleneck. In contrast, deep optimization methods leverage learning-based nonlinear transformations for clustering. They primarily achieve data imputation through the generalization ability of deep models, but their model interpretability may be limited by the black-box nature. Moreover, most existing methods only explore the structure of each view independently, where these structures are fixed and cannot form a complete unified structure. To address these issues, we propose a Structural Optimization-inspired Interpretable Network (SOI-Net) for incomplete multi-view clustering. Specifically, we project the features of all views into a unified representation space with the un-missing information of the views as constraints. By optimizing consistent structural information, we preserve the structures of missing modalities in the unified representation space, thereby mitigating the impact of missing data. Meanwhile, we derive network components based on the optimization problem to guide the learning of structure and representation. The practical significance of these network components provides model design-level interpretability. Extensive experiments on six datasets validate the effectiveness of SOI-Net in handling incomplete multi-view clustering task.

Multi-view representation learning is recognized for its effectiveness in multi-source data analysis, yet it faces significant challenges: 1) Deep model structures remain opaque, lacking interpretability; 2) Research on compatibility models toward multi-feature and multi-relationship data is insufficient. In this paper, we introduce an interpretable multi-view representation learning framework specifically designed for the complex multi view scene. The barrier to achieving compatibility stems from the need to simultaneously process the homogeneous information inherent in multiple features and the heterogeneous characteristic of multi-relation data. To address this, we design an objective function solved by iterative methods to learn comprehensive relationships and the consistent representation. The introduction of comprehensive relationships aims to mitigate mutual interference among different data types while combining information abstracted from original features and relationships into a unified representation. We then convert iterative solutions into feed-forward network layers with embedded learnable modules, resulting in a deep network architecture that is interpretable at the design level. Extensive experimental results demonstrate the superior performance of the proposed method over state-of-the art approaches.

Representation learning techniques effectively unveil latent patterns within raw data. However, the learning process is often marred by uncertainties, such as variations in data quality and heterogeneous scenarios, which greatly affect the reliability of representation learning. In this article, we introduce a reliable representation learning framework to establish a connection between data attributes and modeling strategies, namely the interpretable attribute-oriented representation learning framework. First, by focusing on the inherent knowledge embedded in the data, we decouple it into four principal attributes: fidelity, topology, invariance, and discriminability. To explicitly address these attributes, we incorporate them into an optimization-derived framework using corresponding general loss terms. Furthermore, by treating the iterative solution process as a bridge, each derived network module possesses traceable interpretability, thus laying a reliable foundation. Ultimately, we extend the proposed framework to multisource heterogeneous scenarios, enabling it to adapt to complex environments while maintaining reliability. In essence, our work aims to seamlessly integrate deep representations with prior knowledge during the learning process, thereby creating a solid basis for dependable modeling. Networks derived from the proposed framework achieve promising results, particularly in complex multisource heterogeneous environments, demonstrating both their effectiveness and reliability. The code is available at https://github.com/ZihanFang11/2025_AORLNet_TNNLS.

The rapid evolution of multimedia technology has revolutionized human perception, paving the way for multi-view learning. However, traditional multi-view learning approaches are tailored for scenarios with fixed data views, falling short of emulating the intricate cognitive procedures of the human brain processing signals sequentially. Our cerebral architecture seamlessly integrates sequential data through intricate feed-forward and feedback mechanisms. In stark contrast, traditional methods struggle to generalize effectively when confronted with data spanning diverse domains, highlighting the need for innovative strategies that can mimic the brain's adaptability and dynamic integration capabilities. In this paper, we propose a bio-neurologically inspired multi-view incremental framework named MVIL aimed at emulating the brain's fine-grained fusion of sequentially arriving views. MVIL lies two fundamental modules: structured Hebbian plasticity and synaptic partition learning. The structured Hebbian plasticity reshapes the structure of weights to express the high correlation between view representations, facilitating a fine-grained fusion of view representations. Moreover, synaptic partition learning is efficient in alleviating drastic changes in weights and also retaining old knowledge by inhibiting partial synapses. These modules bionically play a central role in reinforcing crucial associations between newly acquired information and existing knowledge repositories, thereby enhancing the network's capacity for generalization. Experimental results on six benchmark datasets show MVIL's effectiveness over state-of-the-art methods.

Multi-view learning methods leverage multiple data sources to enhance perception by mining correlations across views, typically relying on predefined categories. However, deploying these models in real-world scenarios presents two primary openness challenges. 1) Lack of Interpretability: The integration mechanisms of multi-view data in existing black-box models remain poorly explained; 2) Insufficient Generalization: Most models are not adapted to multi-view scenarios involving unknown categories. To address these challenges, we propose OpenViewer, an openness-aware multi-view learning framework with theoretical support. This framework begins with a Pseudo-Unknown Sample Generation Mechanism to efficiently simulate open multi-view environments and previously adapt to potential unknown samples. Subsequently, we introduce an Expression-Enhanced Deep Unfolding Network to intuitively promote interpretability by systematically constructing functional prior-mapping modules and effectively providing a more transparent integration mechanism for multi-view data. Additionally, we establish a Perception-Augmented Open-Set Training Regime to significantly enhance generalization by precisely boosting confidences for known categories and carefully suppressing inappropriate confidences for unknown ones. Experimental results demonstrate that OpenViewer effectively addresses openness challenges while ensuring recognition performance for both known and unknown samples.

Text-Attributed Graphs (TAGs) are vital for modeling entity relationships across various domains. Graph Neural Networks have become cornerstone for processing graph structures, while the integration of text attributes remains a prominent research. The development of Large Language Models (LLMs) provides new opportunities for advancing textual encoding in TAGs. However, LLMs face challenges in specialized domains due to their limited task-specific knowledge, and fine-tuning them for specific tasks demands significant resources. To cope with the above challenges, we propose HiTuner, a novel framework that leverages fine-tuned Pre-trained Language Models (PLMs) with domain expertise as tuner to enhance the hierarchical LLM contextualized representations for modeling TAGs. Specifically, we first strategically select hierarchical hidden states of LLM to form a set of diverse and complementary descriptions as input for the sparse projection operator. Concurrently, a hybrid representation learning is developed to amalgamate the broad linguistic comprehension of LLMs with task-specific insights of the fine-tuned PLMs. Finally, HiTuner employs a confidence network to adaptively fuse the semantically-augmented representations. Empirical results across benchmark datasets spanning various domains validate the effectiveness of the proposed framework. Our codes are available at: https://github.com/ZihanFang11/HiTuner

Deep anchor-based multi-view clustering methods enhance the scalability of neural networks by utilizing representative anchors to reduce the computational complexity of large-scale clustering. Despite their scalability advantages, existing approaches often incorporate anchor structures in a heuristic or task-agnostic manner, either through post-hoc graph construction or as auxiliary components for message passing. Such designs overlook the core structural demands of anchor-based clustering, neglecting key optimization principles. To bridge this gap, we revisit the underlying optimization problem of large-scale anchor-based multi-view clustering and unfold its iterative solution into a novel deep network architecture, termed LargeMvC-Net. The proposed model decomposes the anchor-based clustering process into three modules: RepresentModule, NoiseModule, and AnchorModule, corresponding to representation learning, noise suppression, and anchor estimation. Each module is derived by unfolding a step of the original optimization procedure into a dedicated network component, providing structural clarity and optimization traceability. In addition, an unsupervised reconstruction loss aligns each view with the anchor-induced latent space, encouraging consistent clustering structures across views. Extensive experiments on several large-scale multi-view benchmarks show that LargeMvC-Net consistently outperforms state-of-the-art methods in terms of both effectiveness and scalability. The source data, code. https://github.com/dushide/LargeMvC-Net_ACMMM_2025, and extended version http://arxiv.org/abs/2507.20980 are available.

Existing multi-view learning models struggle in open-set scenarios due to their implicit assumption of class completeness. Moreover, static view-induced biases, which arise from spurious view-label associations formed during training, further degrade their ability to recognize unknown categories. In this paper, we propose a multi-view open-set learning framework via ambiguity uncertainty calibration and view-wise debiasing. To simulate ambiguous samples, we design O-Mix, a novel synthesis strategy to generate virtual samples with calibrated open-set ambiguity uncertainty. These samples are further processed by an auxiliary ambiguity perception network that captures atypical patterns for improved open-set adaptation. Furthermore, we incorporate an HSIC-based contrastive debiasing module that enforces independence between view-specific ambiguous and view-consistent representations, encouraging the model to learn generalizable features. Extensive experiments on diverse multi-view benchmarks demonstrate that the proposed framework consistently enhances unknown-class recognition while preserving strong closed-set performance. The source code are available https://github.com/ZihanFang11/2025_MOCD_ACMMM

Deep multi-view graph clustering seeks to integrate diverse graph feature sets and uncover consistent information across multiple views. While extensive prior research has utilized various neural network architectures to address multi-view graph clustering challenges, these approaches exhibit notable limitations: 1) The ''black-box'' nature of deep learning models, which obscures their internal mechanisms and impedes interpretability; 2) Insufficient efforts aim to capture low-dimensional representations through graphs that reflect intuitive clustering structures and reduce computational cost. To address these limitations, this paper introduces an interpretable multi-view graph clustering framework constructed with optimization-inspired modules. The proposed approach formulates low-dimensional clustering representation learning from graph matrices as an optimization problem, deriving an iterative solution rooted in this formulation. By seamlessly bridging this optimization process to a deep network architecture, the model learns a low-dimensional clustering representation for graph-structured data across multiple views while adhering to the iterative optimization principles and reducing computational costs. This transparent network design enhances the interpretability of multi-view clustering, enabling intuitive and human-understandable learning of clustering structures. Extensive experimental evaluations validate the proposed framework's superiority over state-of-the-art methods in multi-view clustering tasks while ensuring interpretability and reducing computational costs.

In practical applications, the difficulty of multi-view data annotation poses a challenge for multi-view semi-supervised learning. Although some graph-based approaches have been proposed for this task, they often struggle with capturing long-range information and memory bottlenecks, and usually encounter over-smoothing. To address these issues, this paper proposes an implicit model, named multi-channel Equilibrium Graph Neural Network (MEGNN). Through an equilibrium point iterative process, the proposed MEGNN naturally captures long-range information and effectively reduces the consumption of memory compared with explicit models. Furthermore, the proposed method deals with the issue of over-smoothing in deep graph convolutional networks by residual connection and shrinkage factor. We analyze the effect of the shrinkage factor on the information capturing capability of the model, and demonstrate that the proposed method does not encounter over-smoothing. Comprehensive experimental results demonstrate that the proposed method outperforms the state-of-the-art methods.

Active learning, as a technique, aims to effectively label specific data points while operating within a designated query budget. Nevertheless, the majority of unsupervised active learning algorithms are based on shallow linear representation and lack sufficient interpretability. Furthermore, certain diversity-based methods face challenges in selecting samples that adequately represent the entire data distribution. Inspired by these reasons, in this paper, we propose an unsupervised active learning method on orthogonal projections to construct a deep neural network model. By optimizing the orthogonal projection process, we establish the connection between projection and active learning, consequently enhancing the interpretability of the proposed method. The proposed method can efficiently project the feature space onto a spanned subspace, deriving an indicator matrix while calculating the projection loss. Moreover, we consider the redundancy among samples to ensure both data point diversity and enhancement of clustering-based algorithms. Through extensive comparative experiments on six public datasets, the results demonstrate that the proposed method can effectively select more informative and representative samples and improve performance by up to 11%.

Deep learning, as an important branch of machine learning, has been widely applied in computer vision, natural language processing, speech recognition, and more. However, recent studies have revealed that deep learning systems are vulnerable to backdoor attacks. Backdoor attackers inject a hidden backdoor into the deep learning model, such that the predictions of the infected model will be maliciously changed if the hidden backdoor is activated by input with a backdoor trigger while behaving normally on any benign sample. This kind of attack can potentially result in severe consequences in the real world. Therefore, research on defending against backdoor attacks has emerged rapidly. In this article, we have provided a comprehensive survey of backdoor attacks, detections, and defenses previously demonstrated on deep learning. We have investigated widely used model architectures, benchmark datasets, and metrics in backdoor research and have classified attacks, detections and defenses based on different criteria. Furthermore, we have analyzed some limitations in existing methods and, based on this, pointed out several promising future research directions. Through this survey, beginners can gain a preliminary understanding of backdoor attacks and defenses. Furthermore, we anticipate that this work will provide new perspectives and inspire extra research into the backdoor attack and defense methods in deep learning.

Traditional multiview clustering relies on manually-designed optimization problems based on prior interpretable knowledge to cluster objects with similar attributes, but it may be hindered by limited feature extraction capability. In contrast, deep multiview clustering overcomes this limitation by utilizing learning-based nonlinear transformations for clustering, but it may be restricted by model interpretability caused by blackbox networks. Besides, previous multiview clustering methods only consider homogeneous or heterogeneous situations, resulting in restricted scalability and generalizability. To address the aforementioned issues, we design a universal interpretable clustering framework that accommodates both homogeneous and heterogeneous multiview scenarios. To realize this purpose: 1) we revisit the interpretable knowledge-driven design architecture of traditional multiview methods and formulate clustering optimization problems on multiview data attributes in homogeneous scenarios; 2) the optimization problem is leveraged to derive network modules that learn shared and self-expressive representations for clustering, with the practical meaning of each network component providing model design-level interpretability; 3) the proposed method is extended from homogeneous to heterogeneous scenarios, enhancing its universality for a broader spectrum of multiview clustering tasks; and 4) tailored training loss for the clustering task is employed to inversely enhance the affinity between objects with similar attributes. Extensive experimental results on both homogeneous and heterogeneous multiview datasets demonstrate the superior effectiveness and adaptability of the proposed framework compared to state-of-the-art clustering methods.

Correspondence pruning aims to identify inliers from correspondences severely disturbed by outliers. Although Transformers and graph neural networks have shown impressive results in this field, they are either limited by a narrow receptive field or encounter quadratic computational complexity. To tackle this challenge, this work pioneers the integration of state space model into correspondence pruning task, proposing a Mamba-based framework named MambaMatch. Specifically, to address the limitations of the Mamba architecture in local consensus modeling, we proposes a multi-scale scanning strategy. It first employs an adaptive clustering algorithm to map origin correspondences into spatially coherent feature clusters, constructing a dual-representation space encompassing both full-scale and clustered-scale features. Bidirectional scan operations are then performed at both scales: 1) full-scale scan preserves global structural context, and 2) clustered-scale scan enhances local consistency. Subsequently, a Multi-Scale Interaction layer is designed to dynamically fuse dual-scale features via a cross-attention mechanism, further integrated with a Gated Feed-Forward Network to significantly improve the network's feature discrimination capability. Extensive experiments validate that MambaMatch surpasses state-of-the-art approaches across multiple benchmarks for two-view geometry estimation. Furthermore, MambaMatch exhibits robust generalization across diverse scenarios, tasks, and feature extractors. The source code is available at: https://github.com/mxyttkx/MambaMatch.

Existing methods based on graph convolutional network often struggle with large-scale graphs due to their high computing consumption and inefficiency. Although strategies such as edge sparsification and node sampling can indeed decrease the complexity, they frequently result in information loss and local information bias. Furthermore, in multi-view scenarios, traditional multi-view fusion methods are unable to simultaneously account for both inter-view consistency and intra-view diversity, thus constraining model performance. In this paper, we propose a multi-view block-wise graph convolutional network that effectively addresses the challenges posed by large-scale graphs while exploiting the complementary nature of multi-view information. Specifically, we implement a node segmentation module to partition nodes into view-specific subsets, thereby diminishing computational complexity while preserving local structural information. To enhance feature extraction, plentiful subgraph representations are captured within blocks by alternating graph convolution with graph structure learning under a shared-weight strategy. Finally, the global fusion module introduces a cross-view inter-block loss that progressively aligns block representations across views, alleviates over-smoothing, and yields a consistent and comprehensive common representation. Extensive experiments on diverse large-scale graph datasets demonstrate that the proposed method not only outperforms state-of-the-art approaches in multi-view semi-supervised classification but also exhibits superior scalability and memory efficiency.

Although Graph Neural Networks (GNNs) have exhibited the powerful ability to gather graph-structured information from neighborhood nodes via various message-passing mechanisms, the performance of GNNs is limited by poor generalization and fragile robustness caused by noisy and redundant graph data. As a prominent solution, Graph Augmentation Learning (GAL) has recently received increasing attention in the literature. Among the existing GAL approaches, edge-dropping methods that randomly remove edges from a graph during training are effective techniques to improve the robustness of GNNs. However, randomly dropping edges often results in bypassing critical edges. Consequently, the effectiveness of message passing is weakened. In this paper, we propose a novel adversarial edge-dropping method (ADEdgeDrop) that leverages an adversarial edge predictor guiding the removal of edges, which can be flexibly incorporated into diverse GNN backbones. Employing an adversarial training framework, the edge predictor utilizes the line graph transformed from the original graph to estimate the edges to be dropped, which improves the interpretability of the edge-dropping method. The proposed ADEdgeDrop is optimized alternately by stochastic gradient descent and projected gradient descent. Comprehensive experiments on eight graph benchmark datasets demonstrate that the proposed ADEdgeDrop outperforms state-of-the-art baselines across various GNN backbones, demonstrating improved generalization and robustness.

The challenges posed by heterogeneous data in practical applications have made multiview semi-supervised classification a focus of attention for researchers. While several graph-based approaches have been suggested for this task, they tend to use homogeneous feature propagation, leading to even diffusion of node information to their neighbors. However, this diffusion strategy results in nodes acquiring information of equal proportion from dissimilar samples. In this article, we propose a solution to address these issues by introducing a graph diffusion-induced network for multiview semi-supervised classification. By formulating a discretized partial differential equation on a manifold, we derive a nonlinear and inhomogeneous diffusion equation to govern information propagation on the graph. Then, we investigate the impact of various nonlinear activation functions on random switching edge directions and their suppressive effects on information diffusion between different nodes. In addition, the cross-view consistency under the semi-supervised scenarios is defined and guaranteed for better information fusion. The comprehensive experimental results demonstrate the superiority of the proposed method compared with state-of-the-art approaches. The effectiveness of the proposed approach in handling diverse and heterogeneous data showcases its potential for advancing multiview semi-supervised classification techniques.

Graph convolutional network (GCN) has gained widespread attention in semisupervised classification tasks. Recent studies show that GCN-based methods have achieved decent performance in numerous fields. However, most of the existing methods generally adopted a fixed graph that cannot dynamically capture both local and global relationships. This is because the hidden and important relationships may not be directed exhibited in the fixed structure, causing the degraded performance of semisupervised classification tasks. Moreover, the missing and noisy data yielded by the fixed graph may result in wrong connections, thereby disturbing the representation learning process. To cope with these issues, this article proposes a learnable GCN-based framework, aiming to obtain the optimal graph structures by jointly integrating graph learning and feature propagation in a unified network. Besides, to capture the optimal graph representations, this article designs dual-GCN-based meta-channels to simultaneously explore local and global relations during the training process. To minimize the interference of the noisy data, a semisupervised graph information bottleneck (SGIB) is introduced to conduct the graph structural learning (GSL) for acquiring the minimal sufficient representations. Concretely, SGIB aims to maximize the mutual information of both the same and different meta-channels by designing the constraints between them, thereby improving the node classification performance in the downstream tasks. Extensive experimental results on real-world datasets demonstrate the robustness of the proposed model, which outperforms state-of-the-art methods with fixed-structure graphs.

In the dynamic environment of multimedia-sharing platforms like X (formerly known as Twitter) and TikTok, multimedia recommendation systems have been widely used to help users discover items of interest. However, traditional approaches often fall short, when the item modalities are incomplete, a common issue in real-world scenarios. To this end, we introduce the unified heterogeneous Hypergraph construction for the Incomplete multimedia REcommendation ( HIRE ), a novel framework designed to jointly learn a heterogeneous hypergraph and perform accurate recommendations under incomplete scenarios. HIRE first initializes the unified heterogeneous hypergraph for modality completion and employs self-supervised learning aligned with the contrastive text-centered view for multimedia recommendation. Such integration effectively handles the challenges posed by incomplete modalities, leading to improved recommendation accuracy. Furthermore, we find that the hypergraph directly learned from the HIRE is a dense structure which can be inaccurate and coarse. Therefore, we devise the HIRE framework with Sparse constraint named HIRES , which uniquely integrates optimal transport and a \(\ell_{2,1}\) -norm to refine the hypergraph structure. Our extensive experiments across various datasets demonstrate the superiority of HIRES in addressing incomplete modalities, establishing it as a powerful tool for personalized multimedia recommendations.

Most of existing correspondence pruning methods only concentrate on gathering the context information as much as possible while neglecting effective ways to utilize such information. In order to tackle this dilemma, in this paper we propose Graph Context Transformation Network (GCT-Net) enhancing context information to conduct consensus guidance for progressive correspondence pruning. Specifically, we design the Graph Context Enhance Transformer which first generates the graph network and then transforms it into multi-branch graph contexts. Moreover, it employs self-attention and cross-attention to magnify characteristics of each graph context for emphasizing the unique as well as shared essential information. To further apply the recalibrated graph contexts to the global domain, we propose the Graph Context Guidance Transformer. This module adopts a confident-based sampling strategy to temporarily screen high-confidence vertices for guiding accurate classification by searching global consensus between screened vertices and remaining ones. The extensive experimental results on outlier removal and relative pose estimation clearly demonstrate the superior performance of GCT-Net compared to state-of-the-art methods across outdoor and indoor datasets.

Correspondence pruning aims to establish reliable correspondences between two related images and recover relative camera motion. Existing approaches often employ a progressive strategy to handle the local and global contexts, with a prominent emphasis on transitioning from local to global, resulting in the neglect of interactions between different contexts. To tackle this issue, we propose a parallel context learning strategy that involves acquiring bilateral consensus for the two-view correspondence pruning task. In our approach, we design a distinctive self-attention block to capture global context and parallel process it with the established local context learning module, which enables us to simultaneously capture both local and global consensuses. By combining these local and global consensuses, we derive the required bilateral consensus. We also design a recalibration block, reducing the influence of erroneous consensus information and enhancing the robustness of the model. The culmination of our efforts is the Bilateral Consensus Learning Network (BCLNet), which efficiently estimates camera pose and identifies inliers (true correspondences). Extensive experiments results demonstrate that our network not only surpasses state-of-the-art methods on benchmark datasets but also showcases robust generalization abilities across various feature extraction techniques. Noteworthily, BCLNet obtains significant improvement gains over the second best method on unknown outdoor dataset, and obviously accelerates model training speed.

The sparse interactions between users and items have aggravated the difficulty of their representations in recommender systems. Existing methods leverage tags to alleviate the sparsity problem but ignore prevalent logical relations among items and tags (e.g., membership, hierarchy, and exclusion), which can be leveraged to enhance the accuracy of modeling user preferences and conducting recommendations. To this end, we propose to extract logical relations among item tags from existing tag taxonomies and exploit the individual strengths of the Poincaré and the Lorentz models in hyperbolic space for logical relation modeling towards enhanced recommendations. Moreover, we find that the logical relations directly extracted from existing tag taxonomies can be inaccurate and coarse. Therefore, we further devise innovative consistency-based and granularity- based weighting mechanisms based on user behavior patterns for data-driven logical relation mining that can be jointly optimized along with recommendations in an end-to-end fashion. Extensive experiments on four real-world benchmark datasets show drastic performance gains brought by our proposed framework, which constantly achieves an average of 8.25% improvement over state-of-the-art competitors regarding both Recall and NDCG metrics. Insightful case studies further demonstrate that our automatically refined logical relations are highly accurate and interpretable.

Multimedia recommender systems (MRS) become prevalent due to their rich multimodal data (e.g., visual and textual content). Recent advancements have leveraged Graph Neural Networks (GNNs) to integrate these data, they often fall short in capturing the complex high-order relations within multimodal data, but readily hypergraph structures are not always available. To this end, we introduce the HMRec framework, a novel approach in Heterogeneous Hypergraph Structure Learning tailored for MRS. Specifically, we formulate the construction of a heterogeneous hypergraph as determining item associations across modalities, and introduce an adaptive hypergraph convolution mechanism for differentially weighting multimodal hyperedges. Furthermore, we propose an enhanced multimedia recommendation module, which introduces a contrastive fusion mechanism to effectively integrate graph-view, hypergraph-view, and ID-specific embeddings. Extensive experiments on real-world multimodal datasets show the superiority of our proposed HMRec framework in offering great potential for multimedia recommendations over the state-of-the-art baselines regarding the Recall and NDCG metrics.

The emerging Graph Convolutional Networks (GCNs) have attracted widespread attention in graph learning, due to their good ability of aggregating the information between higher-order neighbors. However, real-world graph data contains high noise and redundancy, making it hard for GCNs to accurately depict the complete relationships between nodes, which seriously degrades the quality of graph representations. Moreover, existing studies commonly ignore the distribution difference between feature and semantic spaces in graphs, causing inferior model generalization. To address these challenges, we propose DIB-RGCN, a novel robust GCN framework, to explore the optimal graph representation with the guidance of the well-designed dual information bottleneck principle. First, we analyze the reasons for distribution differences and theoretically prove that minimal sufficient representations in specific spaces cannot promise optimal performance for downstream tasks. Next, we design new dual channels to regularize feature and semantic spaces, eliminating the sharing of task-irrelevant information between spaces. Different from existing denoising algorithms that adopt a random dropping manner, we innovatively replace potential noisy features and edges with local neighboring representations. This design lowers edge-specific coefficient assignment, alleviating the interference of original representations while retaining graph structures. Further, we maximize the sharing of task-relevant information between feature and semantic spaces to alleviate the difference between them. Using real-world datasets, extensive experiments demonstrate the robustness of the proposed DIB-RGCN, which outperforms state-of-the-art methods on classification tasks.

The inherent variability and unpredictability in open multi-view learning scenarios infuse considerable ambiguity into the learning and decision-making processes of predictors. This demands that predictors not only recognize familiar patterns but also adaptively interpret unknown ones out of training scope. To address this challenge, we propose an Ambiguity-Aware Multi-view Learning Framework, which integrates four synergistic modules into an end-to-end framework to achieve generalizability and reliability beyond the known. By introducing the mixed samples to broaden the learning sample space, accompanied by corresponding soft labels to encapsulate their inherent uncertainty, the proposed method adapts to the distribution of potentially unknown samples in advance. Furthermore, an instance-level sparse inference is implemented to learn sparse approximated points in the multiple view embedding space, and individual view representations are gated by view-level confidence mappings. Finally, a multi-view consistent representation is obtained by dynamically assigning weights based on the degree of cluster-level dispersion. Extensive experiments demonstrate that our approach is effective and stable compared with other state-of-the-art methods in open-world recognition situations.

Facing the increasing heterogeneity of data in the real world, multi-view learning has become a crucial area of research. Graph Convolutional Networks (GCNs) are powerful for modeling both graph structures and features, making them a focal point in multi-view learning research. However, these methods typically only account for static data dependencies within each view separately when constructing the topology necessary for GCNs, overlooking potential relationships across views in multi-view data. Furthermore, there is a notable absence of theoretical guidance for constructing multi-view data topologies, leading to uncertainty regarding the progression of graph embeddings toward a consistent state. To tackle these challenges, we introduce a framework named energy-constrained multi-view graph diffusion. This approach establishes a mathematical correspondence between multi-view data and GCNs via graph diffusion. It treats multi-view data as a unified entity and devises a feature propagation process with inter-view awareness by considering both inter-view and intra-view feature flow across the entire system. Additionally, an energy function is introduced to guide the inter- and intra-view diffusion, ensuring that the representations converge towards global consistency. The empirical research on several benchmark datasets substantiates the benefits of the proposed method.

Graph Convolutional Network (GCN) has drawn widespread attention in data mining on graphs due to its outstanding performance and rigor theoretical guarantee. However, some recent studies have revealed that GCN-based methods may mine latent information insufficiently owing to the underutilization of the feature space . Besides, the unlearnable topology also significantly imperils the performance of GCN-based methods. In this paper, we conduct experiments to investigate these issues, finding that GCN does not fully consider the potential structure in the feature space, and a fixed topology deteriorates the robustness of GCN. Thus, it is desired to distill node features and establish a learnable graph. Motivated by this goal, we propose a framework dubbed G raph C onvolutional N etwork with e lastic t opology (GCNet 1 ). With the analysis of the optimization for the proposed flexible Laplacian embedding, GCNet is naturally constructed by alternative graph convolutional layers and adaptive topology learning layers. GCNet aims to deeply explore the feature space and employ the mined information to construct a learnable topology, which leads to a more robust graph representation. In addition, a set-level orthogonal loss is utilized to meet the orthogonal constraint required by the flexible Laplacian embedding and promote better class separability . Moreover, comprehensive experiments indicate that GCNet achieves remarkable performance and generalization on several real-world datasets.

Biomedical relation extraction aims to identify underlying relationships among entities, such as gene associations and drug interactions, within biomedical texts. Despite advancements in relation extraction in general knowledge domains, the scarcity of labeled training data remains a significant challenge in the biomedical field. This paper provides a novel approach for biomedical relation extraction that leverages a noisy student self-training strategy combined with negative learning. This method addresses the challenge of data insufficiency by utilizing distantly supervised data to generate high-quality labeled samples. Negative learning, as opposed to traditional positive learning, offers a more robust mechanism to discern and relabel noisy samples, preventing model overfitting. The integration of these techniques ensures enhanced noise reduction and relabeling capabilities, leading to improved performance even with noisy datasets. Experimental results demonstrate the effectiveness of the proposed framework in mitigating the impact of noisy data and outperforming existing benchmarks.

Graph data have become increasingly important, and graph node clustering has emerged as a fundamental task in data analysis. In recent years, graph node clustering has gradually moved from traditional shallow methods to deep neural networks due to the powerful representation capabilities of deep learning. In this article, we review some representatives of the latest graph node clustering methods, which are classified into three categories depending on their principles. Extensive experiments are conducted on real-world graph datasets to evaluate the performance of these methods. Four mainstream evaluation performance metrics are used, including clustering accuracy, normalized mutual information, adjusted rand index, and F1-score. Based on the experimental results, several potential research challenges and directions in the field of deep graph node clustering are pointed out. This work is expected to facilitate researchers interested in this field to provide some insights and further promote the development of deep graph node clustering.

In recent years, reconstructing features and learning node representations by graph autoencoders (GAE) have attracted much attention in deep graph node clustering. However, existing works often overemphasize structural information and overlook the impact of real-world prevalent noise on feature learning and clustering with graph data, which may be detrimental to robust training. To address these issues, the utilization of a masking strategy that specifically focuses on feature reconstruction may mitigate these limitations. In this article, we propose a graph node clustering generative method named deep masked graph node clustering (DMGNC), which leverages a masked autoencoder to effectively reconstruct node features, enabling the discovery of latent information crucial for accurate node clustering. Additionally, a clustering self-optimization module is designed to guide the iterative update of our end-to-end clustering framework. Further, we extend the masked graph autoencoder (MGA) and develop a contrastive method called deep masked graph node contrastive clustering (DMGNCC), which applies the MGA to graph node contrastive learning at both the node level and the class level in a united model. Extensive experimental results on real-world graph benchmark datasets demonstrate the effectiveness and superiority of the proposed method.

Existing multi-view graph learning methods often rely on consistent information for similar nodes within and across views, however they may lack adaptability when facing diversity challenges from noise, varied views, and complex data distributions. These challenges can be mainly categorized into: 1) View-specific diversity within intra-view from noise and incomplete information; 2) Cross-view diversity within inter-view caused by various latent semantics; 3) Cross-group diversity within inter-group due to data distribution differences. To this end, we propose a universal multi-view consensus graph learning framework that considers both original and generative graphs to balance consistency and diversity. Specifically, the proposed framework can be divided into the following four modules: i) Multi-channel graph module to extract principal node information, ensuring view-specific and cross-view consistency while mitigating view-specific and cross-view diversity within original graphs; ii) Generative module to produce cleaner and more realistic graphs, enriching graph structure while maintaining view-specific consistency and suppressing view-specific diversity; iii) Contrastive module to collaborate on generative semantics to facilitate cross-view consistency and reducing cross-view diversity within generative graphs; iv) Consensus graph module to consolidate learning a consensual graph, pursuing cross-group consistency and cross-group diversity. Extensive experimental results on real-world datasets demonstrate its effectiveness and superiority.

Multi-view data containing complementary and consensus information can facilitate representation learning by exploiting the intact integration of multi-view features. Because most objects in the real world often have underlying connections, organizing multi-view data as heterogeneous graphs is beneficial to extracting latent information among different objects. Due to the powerful capability to gather information of neighborhood nodes, in this article, we apply Graph Convolutional Network (GCN) to cope with heterogeneous graph data originating from multi-view data, which is still under-explored in the field of GCN. In order to improve the quality of network topology and alleviate the interference of noises yielded by graph fusion, some methods undertake sorting operations before the graph convolution procedure. These GCN-based methods generally sort and select the most confident neighborhood nodes for each vertex, such as picking the top- k nodes according to pre-defined confidence values. Nonetheless, this is problematic due to the non-differentiable sorting operators and inflexible graph embedding learning, which may result in blocked gradient computations and undesired performance. To cope with these issues, we propose a joint framework dubbed Multi-view Graph Convolutional Network with Differentiable Node Selection (MGCN-DNS), which is constituted of an adaptive graph fusion layer, a graph learning module, and a differentiable node selection schema. MGCN-DNS accepts multi-channel graph-structural data as inputs and aims to learn more robust graph fusion through a differentiable neural network. The effectiveness of the proposed method is verified by rigorous comparisons with considerable state-of-the-art approaches in terms of multi-view semi-supervised classification tasks, and the experimental results indicate that MGCN-DNS achieves pleasurable performance on several benchmark multi-view datasets.

Deep learning-based clustering methods, especially those incorporating deep generative models, have recently shown noticeable improvement on many multimedia benchmark datasets. However, existing generative models still suffer from unstable training, and the gradient vanishes, which results in the inability to learn desirable embedded features for clustering. In this paper, we aim to tackle this problem by exploring the capability of Wasserstein embedding in learning representative embedded features and introducing a new clustering module for jointly optimizing embedding learning and clustering. To this end, we propose Wasserstein embedding clustering (WEC), which integrates robust generative models with clustering. By directly minimizing the discrepancy between the prior and marginal distribution, we transform the optimization problem of Wasserstein distance from the original data space into embedding space, which differs from other generative approaches that optimize in the original data space. Consequently, it naturally allows us to construct a joint optimization framework with the designed clustering module in the embedding layer. Due to the substitutability of the penalty term in Wasserstein embedding, we further propose two types of deep clustering models by selecting different penalty terms. Comparative experiments conducted on nine publicly available multimedia datasets with several state-of-the-art methods demonstrate the effectiveness of our method.

Existing representation learning approaches lie predominantly in designing models empirically without rigorous mathematical guidelines, neglecting interpretation in terms of modeling. In this work, we propose an optimization-derived representation learning network that embraces both interpretation and extensibility. To ensure interpretability at the design level, we adopt a transparent approach in customizing the representation learning network from an optimization perspective. This involves modularly stitching together components to meet specific requirements, enhancing flexibility and generality. Then, we convert the iterative solution of the convex optimization objective into the corresponding feed-forward network layers by embedding learnable modules. These above optimization-derived layers are seamlessly integrated into a deep neural network architecture, allowing for training in an end-to-end fashion. Furthermore, extra view-wise weights are introduced for multi-view learning to discriminate the contributions of representations from different views. The proposed method outperforms several advanced approaches on semi-supervised classification tasks, demonstrating its feasibility and effectiveness.

With the rapid growth of activities on the web, large amounts of interaction data on multimedia platforms are easily accessible, including e-commerce, music sharing, and social media. By discovering various interests of users, recommender systems can improve user satisfaction without accessing overwhelming personal information. Compared to graph-based models, hypergraph-based collaborative filtering has the ability to model higher-order relations besides pair-wise relations among users and items, where the hypergraph structures are mainly obtained from specialized data or external knowledge. However, the above well-constructed hypergraph structures are often not readily available in every situation. To this end, we first propose a novel framework named HGRec, which can enhance recommendation via automatic hypergraph generation. By exploiting the clustering mechanism based on the user/item similarity, we group users and items without additional knowledge for hypergraph structure learning and design a cross-view recommendation module to alleviate the combinatorial gaps between the representations of the local ordinary graph and the global hypergraph. Furthermore, we devise a sparse optimization strategy to ensure the effectiveness of hypergraph structures, where a novel integration of the$\ell _{2,1}$-norm and optimal transport framework is designed for hypergraph generation. We term the model HGRec with sparse optimization strategy as HGRec++. Extensive experiments on public multi-domain datasets demonstrate the superiority brought by our HGRec++, which gains average 8.1$\%$and 9.8$\%$improvement over state-of-the-art baselines regarding Recall and NDCG metrics, respectively.

Multi-view learning is a promising research field that aims to enhance learning performance by integrating information from diverse data perspectives. Due to the increasing interest in graph neural networks, researchers have gradually incorporated various graph models into multi-view learning. Despite significant progress, current methods face challenges in extracting information from multiple graphs while simultaneously accommodating specific downstream tasks. Additionally, the lack of a subsequent refinement process for the learned graph leads to the incorporation of noise. To address the aforementioned issues, we propose a method named generative essential graph convolutional network for multi-view semi-supervised classification. Our approach integrates the extraction of multi-graph consistency and complementarity, graph refinement, and classification tasks within a comprehensive optimization framework. This is accomplished by extracting a consistent graph from the shared representation, taking into account the complementarity of the original topologies. The learned graph is then optimized through downstream-specific tasks. Finally, we employ a graph convolutional network with a learnable threshold shrinkage function to acquire the graph embedding. Experimental results on benchmark datasets demonstrate the effectiveness of our approach.

Graph convolutional network (GCN) with the powerful capacity to explore graph-structural data has gained noticeable success in recent years. Nonetheless, most of the existing GCN-based models suffer from the notorious over-smoothing issue, owing to which shallow networks are extensively adopted. This may be problematic for complex graph datasets because a deeper GCN should be beneficial to propagating information across remote neighbors. Recent works have devoted effort to addressing over-smoothing problems, including establishing residual connection structure or fusing predictions from multilayer models. Because of the indistinguishable embeddings from deep layers, it is reasonable to generate more reliable predictions before conducting the combination of outputs from various layers. In light of this, we propose an alternating graph-regularized neural network (AGNN) composed of graph convolutional layer (GCL) and graph embedding layer (GEL). GEL is derived from the graph-regularized optimization containing Laplacian embedding term, which can alleviate the over-smoothing problem by periodic projection from the low-order feature space onto the high-order space. With more distinguishable features of distinct layers, an improved Adaboost strategy is utilized to aggregate outputs from each layer, which explores integrated embeddings of multi-hop neighbors. The proposed model is evaluated via a large number of experiments including performance comparison with some multilayer or multi-order graph neural networks, which reveals the superior performance improvement of AGNN compared with the state-of-the-art models.

Due to the powerful capability to gather the information of neighborhood nodes, Graph Convolutional Network (GCN) has become a widely explored hotspot in recent years. As a well-established extension, Graph AutoEncoder (GAE) succeeds in mining underlying node representations via evaluating the quality of adjacency matrix reconstruction from learned features. However, limited works on GAE were devoted to leveraging both semantic and topological graphs, and they only indirectly extracted the relationships between graphs via weights shared by features. To better capture the connections between nodes from these two types of graphs, this paper proposes a graph neural network dubbed Dual Low-Rank Graph AutoEncoder (DLR-GAE), which takes both semantic and topological homophily into consideration. Differing from prior works that share common weights between GCNs, the presented DLR-GAE conducts sustained exploration of low-rank information between two distinct graphs, and reconstructs adjacency matrices from learned latent factors and embeddings. In order to obtain valid adjacency matrices that meet certain conditions, we design some surrogates and projections to restrict the learned factor matrix. We compare the proposed model with state-of-the-art methods on several datasets, which demonstrates the superior accuracy of DLR-GAE in semi-supervised classification.

Graph convolutional networks (GCNs) have been attracting widespread attentions due to their encouraging performance and powerful generalizations. However, few work provide a general view to interpret various GCNs and guide GCNs' designs. In this paper, by revisiting the original GCN, we induce an interpretable regularizer-centerd optimization framework, in which by building appropriate regularizers we can interpret most GCNs, such as APPNP, JKNet, DAGNN, and GNN-LF/HF. Further, under the proposed framework, we devise a dual-regularizer graph convolutional network (dubbed tsGCN) to capture topological and semantic structures from graph data. Since the derived learning rule for tsGCN contains an inverse of a large matrix and thus is time-consuming, we leverage the Woodbury matrix identity and low-rank approximation tricks to successfully decrease the high computational complexity of computing infinite-order graph convolutions. Extensive experiments on eight public datasets demonstrate that tsGCN achieves superior performance against quite a few state-of-the-art competitors w.r.t. classification tasks.

Multi-omics data provides a wealth of information concerning disease mechanisms, which benefits the exploration of the intricate molecular phenomena underlying diseases. In recent years, considerable endeavors have been directed towards the combination of graph convolutional network, which has the powerful ability to gather information, with multi-omics learning methods to obtain more reliable results. For achieving this pursuit, an essential challenge is data integration. Against this backdrop, we propose a unified framework named multi-level knowledge integration with graph convolutional network, which effectively incorporates multiple prior knowledge and omics data to learn an intrinsic representation. In specific, the model consists of two subnetworks: an attribute-level module and a sample-level module. The former firstly aggregates the knowledge given by the prior biological graphs into low-dimensional embeddings, and then maximizes the consistency between these prior views via optimizing a contrastive loss for attaining the attribute-based representations. The latter leverages an encoder to dimensionalize the original multi-omics data to attain more dominant sample knowledge, and subsequently utilizes another contrastive loss to align these representations between multiple omics for learning the global sample-level information. Comprehensive experiments are performed to show that the proposed model surpasses other state-of-the-art methods.

In recent years, deep multi-view representation learning has made considerable achievements due to its excellent nonlinear mapping capability. Yet its development is limited by the challenge of interpreting the underlying structure. For this reason, we propose a differentiable multi-view representation learning network to address the aforementioned issue, which is equipped with the interpretable working mechanism of sparse low-rank decomposition and outstanding representation ability of neural networks. The network is constructed by stacking multiple differentiable blocks that are explicitly reformulated from the optimization objective exhibiting interpretability. Benefiting from end-to-end optimization of deep networks, it can efficiently learn an interpretable deep representation of high-dimensional features from multi-view data. Extensive experimental results on several benchmark multi-view datasets demonstrate the effectiveness of the learned representation in comparison to several state-of-the-art algorithms.

As researchers strive to narrow the gap between machine intelligence and human through the development of artificial intelligence multimedia technologies, it is imperative that we recognize the critical importance of trustworthiness in open-world, which has become ubiquitous in all aspects of daily life for everyone. However, several challenges may create a crisis of trust in current open-world artificial multimedia systems that need to be bridged: 1) Insufficient explanation of predictive results; 2) Inadequate generalization for learning models; 3) Poor adaptability to uncertain environments. Consequently, we explore a neural program to bridge trustworthiness and open-world learning, extending from single-modal to multi-modal scenarios for readers.1) To enhance design-level interpretability, we first customize trustworthy networks with specific physical meanings; 2) We then design environmental well-being task-interfaces via flexible learning regularizers for improving the generalization of trustworthy learning; 3) We propose to increase the robustness of trustworthy learning by integrating open-world recognition losses with agent mechanisms. Eventually, we enhance various trustworthy properties through the establishment of design-level explainability, environmental well-being task-interfaces and open-world recognition programs. As a result, these designed open-world protocols are applicable across a wide range of surroundings, under open-world multimedia recognition scenarios with significant performance improvements observed.

With the rapid growth of multimedia-sharing platforms (e.g. Twitter and TikTok), multimedia recommender systems have become fundamental for helping users alleviate information overload and discover items of interest. Existing multimedia recommendation methods often incorporate various auxiliary modalities (e.g., visual, textual, and acoustic) to describe item characteristics and improve task performance. However, these methods usually assume that each item is associated with complete modalities, ignoring the prevalence of missing modality issues in real-world scenarios. To deal with the challenge of missing modalities, in this paper, we propose a novel framework of Contrastive Intra- and Inter-Modality Generation (CI2MG) for enhancing incomplete multimedia recommendation. We first develop a contrastive intra- and inter-modality generation module for the missing modalities, where the intra-modality representation is updated through clustering-based hypergraph convolution and inter-modality representation is obtained by optimal transport between different modalities. To tackle the challenge of insufficient and incomplete supervision labels during intra- and inter-modality generation, a modality-aware contrastive learning paradigm is introduced based on an augmentation between the intra-modality view and inter-modality view. Furthermore, to learn task-related representations from the generative modalities and further improve the performance of recommendation, we design an enhanced multimedia recommendation module to alleviate the influences driven by task-irrelevant noise. Extensive experiments on real-world datasets show the superiority of our proposed CI2MG framework in offering great potential for personalized multimedia recommendation over the state-of-the-art baselines regarding Recall, NDCG, and Precision metrics.

Unsupervised clustering is a crucial issue in data mining and pattern recognition. Based on deep learning paradigms, deep clustering algorithms have been studied extensively and obtained superior performance in various applications. However, most of previous methods did not use helpful information from neighborhood relations to form group-separated space, and the feature embedding is usually distorted during the training process. To tackle the former limitation, we develop a graph convolution based unsupervised learning algorithm named Stacked Graph Autoencoder (SGAE). Specifically, SGAE utilizes the message passing mechanism to aggregate information from neighbors and obtain a meaningful and separated latent representation. Since the adjacency matrix is unavailable in clustering tasks, a graph construction approach with two pruning strategies is introduced to generate a transition matrix. To reduce the distortion caused by the multi-layered network training process, we further propose a topological structure preservation mechanism. It uses the constructed adjacency graph as supervised information, to maintain the relationship between nodes in the original space. Experiments on several popular benchmark datasets show that SGAE achieves significant improvements compared to unsupervised and semi-supervised deep clustering methods.

Multiview learning aims to learn beneficial patterns from heterogeneous data sources and has captured growing attention in recent years. Most of the previous research studies focused on searching for an effective feature embedding of downstream tasks using diverse optimization algorithms, however, very limited work has been conducted to explore the connection between multiview learning and deep neural networks of structure sharing hidden layers. In this article, we propose a multiview deep matrix factorization model to learn a shared compact representation from multiview data. First, the proposed model constructs a multiview auto-encoder architecture with one shared encoder and multiple decoders, where each view corresponds to a factorization and the shared encoder leads to a common hidden layer. Accordingly, matrix factorizations from multiview data share the last hidden layer for a high-level semantic representation. Second, the nonnegativity constraint of the learned representation is transformed to the projection operation, which can be easily achieved by activating weights of the shared encoder network. Third, this network is trained with a joint loss of the reconstruction error and the compactness loss. By employing the clustering layer, the proposed method serves as an end-to-end multiview clustering method. Finally, comprehensive experiments on nine real-world datasets demonstrate the superiority of the proposed method against state-of-the-art multiview clustering methods.

Establishing reliable correspondences via a deep learning network is an important task in remote sensing, photogrammetry and other computer vision fields. It usually requires mining the relationship among correspondences to aggregate both local and global context. However, current methods are insufficient to effectively acquire context information with high reliability. In this paper, we propose a graph context attention based network (called GCA-Net) to capture and leverage abundant contextual information for feature matching. Specifically, we design a graph context attention block which generates multi-path graph contexts and softly fuses them to combine respective advantages. In addition, for building the graph context containing stronger representation ability and outlier resistance ability, we further design a local-global channel mining block to gather context information by focusing on the significant part as well as to mine dependencies among channels of correspondences in both local and global aspects. The proposed GCA-Net is able to effectively infer the probability of correspondences being inliers or outliers and estimate the essential matrix meanwhile. Extensive experimental results for outlier removal and relative pose estimation demonstrate that GCA-Net outperforms the state-of-art methods on both outdoor and indoor datasets (i.e.,YFCC100M and SUN3D). In addition, experiments extended to remote sensing and point cloud scenes also demonstrate the powerful generalization capability of our network.

With the development of biotechnology, a large amount of multi-omics data have been collected for precision medicine. There exists multiple graph-based prior biological knowledge about omics data, such as gene-gene interaction networks. Recently, there has been an increasing interest in introducing graph neural networks (GNNs) into multi-omics learning. However, existing methods have not fully exploited these graphical priors since none have been able to integrate knowledge from multiple sources simultaneously. To solve this problem, we propose a multi-omics data analysis framework by incorporating multiple prior knowledge into graph neural network (MPK-GNN). To the best of our knowledge, this is the first attempt to introduce multiple prior graphs into multi-omics data analysis. Specifically, the proposed method contains four parts: (1) a feature-level learning module to aggregate information from prior graphs; (2) a projection module to maximize the agreement among prior networks by optimizing a contrastive loss; (3) a sample-level module to learn a global representation from input multi-omics features; (4) a task-specific module to flexibly extend MPK-GNN for various downstream multi-omics analysis tasks. Finally, we verify the effectiveness of the proposed multi-omics learning algorithm on the cancer molecular subtype classification task. Experimental results show that MPK-GNN outperforms other state-of-the-art algorithms, including multi-view learning methods and multi-omics integrative approaches.

The low-rank matrix completion has gained rapidly increasing attention from researchers in recent years for its efficient recovery of the matrix in various fields. Numerous studies have exploited the popular neural networks to yield low-rank outputs under the framework of low-rank matrix factorization. However, due to the discontinuity and nonconvexity of rank function, it is difficult to directly optimize the rank function via back propagation. Although a large number of studies have attempted to find relaxations of the rank function, e.g., Schatten-$p$norm, they still face the following issues when updating parameters via back propagation: 1) These methods or surrogate functions are still non-differentiable, bringing obstacles to deriving the gradients of trainable variables. 2) Most of these surrogate functions perform singular value decomposition upon the original matrix at each iteration, which is time-consuming and blocks the propagation of gradients. To address these problems, in this paper, we develop an efficient block-wise model dubbed differentiable low-rank learning (DLRL) framework that adopts back propagation to optimize the Multi-Schatten-$p$norm Surrogate (MSS) function. Distinct from the original optimization of this surrogate function, the proposed framework avoids singular value decomposition to admit the gradient propagation and builds a block-wise learning scheme to minimize values of Schatten-$p$norms. Accordingly, it speeds up the computation and makes all parameters in the proposed framework learnable according to a predefined loss function. Finally, we conduct substantial experiments in terms of image recovery and collaborative filtering. The experimental results verify the superiority of the proposed framework in terms of both runtimes and learning performance compared with other state-of-the-art low-rank optimization methods. Our codes are available athttps://github.com/chenzl23/DLRL.

Multi-view subspace clustering aims to utilize the comprehensive information of multi-source features to aggregate data into multiple subspaces. Recently, low-rank tensor learning has been applied to multi-view subspace clustering, which explores high-order correlations of multi-view data and has achieved remarkable results. However, these existing methods have certain limitations: 1) The learning processes of low-rank tensor and label indicator matrix are independent. 2) Variable contributions of different views to the consistent clustering results are not discriminated. To handle these issues, we propose a unified framework that integrates low-rank tensor learning and spectral embedding (ULTLSE) for multi-view subspace clustering. Specifically, the proposed model adopts the tensor singular value decomposition (t-SVD) based tensor nuclear norm to encode the low-rank property of the self-representation tensor, and a label indicator matrix via spectral embedding is simultaneously exploited. To distinguish the importance of various views, we learn a quantifiable weighting coefficient for each view. An effective recursion optimization algorithm is also developed to address the proposed model. Finally, we conduct comprehensive experiments on eight real-world datasets with three categories. The experimental results indicate that the proposed ULTLSE is advanced over existing state-of-the-art clustering methods.

The goal of multi-view learning is to learn latent patterns from various data sources. Most of previous research focused on fitting feature embedding in target tasks. There is very limited research on the connection between feature representations with hidden layers of neural networks. In this paper, a multi-view deep matrix factorization model is proposed to learn a shared feature representation. The proposed model automatically explores the most discriminative features of multi-view data and makes these features meet the requirements of specific applications. Here we explore the connection between deep learning and feature representations. First, the model constructs a scalable neural network with shared hidden layers for exploring a low-dimensional representations of all views. Second, the quality of representation matrix is evaluated via relaxed graph regularization and evaluators to improve the feature representation capability of matrix factorization. Finally, the effectiveness of the proposed method is verified through comparative experiments with eight state-of-the-art multi-view clustering algorithms on eight real-world datasets.

Semi-supervised learning is a common way that investigates how to improve performance of a visual learning model, while data annotation is far from sufficient. Recent works in semi-supervised deep learning have successfully applied consistency regularization, which encourages a model to maintain consistent predictions for different perturbed versions of an image. However, most of such methods ignore the category correlation of image features, especially when exploiting strong augmentation methods for unlabeled images. To address this problem, we propose PConMatch, a model that leverages a probabilistic contrastive learning framework to separate the features of strongly-augmented versions from different classes. A semi-supervised probabilistic contrastive loss is designed, which takes both labeled and unlabeled samples into account and develops an auxiliary module to generate a probability score to measure the model prediction confidence for each sample. Specifically, PConMatch first generates a pair of weakly-augmented versions for each labeled sample, and produces a weakly-augmented version and a corresponding pair of strongly-augmented versions for each unlabeled sample. Second, a probability score module is proposed to assign pseudo-labeling confidence scores to strongly-augmented unlabeled images. Finally, the probability score of each sample is further passed to the contrastive loss, combining with consistency regularization to enable the model to learn better feature representations. Extensive experiments on four publicly available image classification benchmarks demonstrate that the proposed approach achieves state-of-the-art performance in image classification. Several rigorous ablation studies are conducted to validate the effectiveness of the method.

As real-world data become increasingly heterogeneous, multi-view semi-supervised learning has garnered widespread attention. Although existing studies have made efforts towards this and achieved decent performance, they are restricted to shallow models and how to mine deeper information from multiple views remains to be investigated. As a recently emerged neural network, Graph Convolutional Network (GCN) exploits graph structure to propagate label signals and has achieved encouraging performance, and it has been widely employed in various fields. Nonetheless, research on solving multi-view learning problems via GCN is limited and lacks interpretability. To address this gap, in this paper we propose a framework termed Interpretable Multi-view Graph Convolutional Network (IMvGCN11Code is available athttps://github.com/ZhihaoWu99/IMvGCN.). We first combine the reconstruction error and Laplacian embedding to formulate a multi-view learning problem that explores the original space from feature and topology perspectives. In light of a series of derivations, we establish a potential connection between GCN and multi-view learning, which holds significance for both domains. Furthermore, we propose an orthogonal normalization method to guarantee the mathematical connection, which solves the intractable problem of orthogonal constraints in deep learning. In addition, the proposed framework is applied to the multi-view semi-supervised learning task. Comprehensive experiments demonstrate the superiority of our proposed method over other state-of-the-art methods.

Deep multi-view representation learning focuses on training a unified low-dimensional representation for data with multiple sources or modalities. With the rapidly growing attention of graph neural networks, more and more researchers have introduced various graph models into multi-view learning. Although considerable achievements have been made, most existing methods usually propagate information in a single view and fuse multiple information only from the perspective of attributes or relationships. To solve the aforementioned problems, we propose an efficient model termed Dual Fusion-Propagation Graph Neural Network (DFP-GNN) and apply it to deep multi-view clustering tasks. The proposed method is designed with three submodules and has the following merits: a) The proposed view-specific and cross-view propagation modules can capture the consistency and complementarity information among multiple views; b) The designed fusion module performs multi-view information fusion with the attributes of nodes and the relationships among them simultaneously. Experiments on popular databases show that DFP-GNN achieves significant results compared with several state-of-the-art algorithms.

Recently deep learning methods have shown significant progress in data clustering tasks. Deep clustering methods (including distance-based methods and subspace-based methods) integrate clustering and feature learning into a unified framework, where there is a mutual promotion between clustering and representation. However, deep subspace clustering methods are usually in the framework of self-expressive model and hence have quadratic time and space complexities, which prevents their applications in large-scale clustering and real-time clustering. In this paper, we propose a new mechanism for deep clustering. We aim to learn the subspace bases from deep representation in an iterative refining manner while the refined subspace bases help learning the representation of the deep neural networks in return. The proposed method is out of the self-expressive framework, scales to the sample size linearly, and is applicable to arbitrarily large datasets and online clustering scenarios. More importantly, the clustering accuracy of the proposed method is much higher than its competitors. Extensive comparison studies with state-of-the-art clustering approaches on benchmark datasets demonstrate the superiority of the proposed method.

Multi-view data often contain redundant information that cannot be simply spliced. Many existing methods for processing them by assigning weights to each view cannot capture features dynamically. Therefore, we propose a multi-view deep matrix factorization method via neural networks that captures semantic hierarchical information of the data and dynamically produces a consistent representation using the complementarity of multi-view features. Due to the usefulness of deep matrix factorization, the generated representation is easily interpretable. The proposed method yields a harmonized representation directly from multi-view data without an extra weight learning process. In addition, we use a multi-path network to search for a consensual solution and obtain an optimal result. Additional feature optimization is used to enhance the discriminative characterization of the representation matrix. Finally, experiments on four real-world datasets show that the proposed method is superior to state-of-the-arts.

Improving the performance of machine learning algorithms to overcome the curse of dimensionality while maintaining interpretability is still a challenging issue for researchers in artificial intelligence. Patch learning (PL), based on the improved adaptive network-based fuzzy inference system (ANFIS) and continuous local optimization for the input domain, is characterized by high accuracy. However, PL can only handle low-dimensional data set regression. Based on the parallel and serial ensembles, two deep patch learning algorithms with embedded adaptive fuzzy systems (DPLFSs) are proposed in this paper. First, using the maximum information coefficient (MIC) and Pearson's correlation coefficients for feature selection, the variables with the least relationship (linear or nonlinear) are excluded. Second, principal component analysis is used to reduce the complexity further of DPLFSs. Meanwhile, fuzzy C-means clustering is used to enhance the interpretability of DPLFSs. Then, an improved PL method is put forward for the training of each sub-fuzzy system in a fashion of bottom-up layer-by-layer, and finally, the structure optimization is performed to significantly improve the interpretability of DPLFSs. Experiments on several benchmark data sets show the advantages of a DPLFS: (1) it can handle medium-scale data sets; (2) it can overcome the curse of dimensionality faced by PL; (3) its precision and generalization are greatly improved; and (4) it can overcome the poor interpretability of deep learning networks. Compared with shallow and deep learning algorithms, DPLFSs have the advantages of interpretability, self-learning, and high precision. DPLFS1 is superior for medium-scale data; DPLFS2 is more efficient and effective for high-dimensional problems, has a faster convergence, and is more interpretable.

Sparsity-constrained optimization problems are common in machine learning, such as sparse coding, low-rank minimization and compressive sensing. However, most of previous studies focused on constructing various hand-crafted sparse regularizers, while little work was devoted to learning adaptive sparse regularizers from given input data for specific tasks. In this paper, we propose a deep sparse regularizer learning model that learns data-driven sparse regularizers adaptively. Via the proximal gradient algorithm, we find that the sparse regularizer learning is equivalent to learning a parameterized activation function. This encourages us to learn sparse regularizers in the deep learning framework. Therefore, we build a neural network composed of multiple blocks, each being differentiable and reusable. All blocks contain learnable piecewise linear activation functions which correspond to the sparse regularizer to be learned. Furthermore, the proposed model is trained with back propagation, and all parameters in this model are learned end-to-end. We apply our framework to multi-view clustering and semi-supervised classification tasks to learn a latent compact representation. Experimental results demonstrate the superiority of the proposed framework over state-of-the-art multi-view learning models.

In the last two decades, object tracking has been one of the prevalent fields in social media. Object tracking uses a dynamic model to track the same target, aiming to analyze the same social object and its behavior in a set of consecutive video frames. In this article, we provide a comprehensive survey on some representative and latest correlation-filter-based object tracking methods and compare their respective strengths and weaknesses under the theoretical and experimental analyses. First of all, we divide these correlation-filter-based object tracking methods into four categories according to their characteristics, including categorized features, space weight factors, scale factors, and expert strategies. Second, extensive experiments on benchmark datasets with 11 tracking challenges are performed and discussed. Finally, this article also attempts to provide some insights for the readers to comprehend and improve the involved tracking algorithms. It is expected that the overview would present a detailed framework and some enlightenments of object tracking to readers who are new in this area.

Recent researches on multiview learning have received widespread attention due to the increasing generalization of multiview data. As an effective probabilistic model, random walk has also shown encouraging performance in various fields. To further exploit the potential of utilizing random walk schemes to address multiview learning problems, this article proposes a simple yet efficient multigraph random walk scheme for both multiview clustering and semisupervised classification tasks. The proposed model integrates random walk with multiview learning, and recursively learns a globally stable probability distribution matrix from multiple views, on the basis of which the label indicator is obtained in the scene of clustering or semisupervised classification. Furthermore, an adaptive weight vector is learned to incorporate the diversity and complementarity of multiview data. Besides, the relationships between the proposed scheme and spectral clustering, neighborhood embedding and manifold embedding are analyzed theoretically. Finally, comprehensive comparative experiments are conducted with several state-of-the-art multiview clustering and semisupervised classification methods on eight real-world datasets. The experimental results demonstrate the superiority of the proposed method in terms of both clustering and classification performance.

In this paper, we propose a novel multi-scale attention based network (called MSA-Net) for feature matching problems. Current deep networks based feature matching methods suffer from limited effectiveness and robustness when applied to different scenarios, due to random distributions of outliers and insufficient information learning. To address this issue, we propose a multi-scale attention block to enhance the robustness to outliers, for improving the representational ability of the feature map. In addition, we also design a novel context channel refine block and a context spatial refine block to mine the information context with less parameters along channel and spatial dimensions, respectively. The proposed MSA-Net is able to effectively infer the probability of correspondences being inliers with less parameters. Extensive experiments on outlier removal and relative pose estimation have shown the performance improvements of our network over current state-of-the-art methods with less parameters on both outdoor and indoor datasets. Notably, our proposed network achieves an 11.7% improvement at error threshold 5° without RANSAC than the state-of-the-art method on relative pose estimation task when trained on YFCC100M dataset.

In tensor-based multi-view learning methods, the self-representation based subspace clustering is widely researched, which is effective but heavy in high computational complexity. Furthermore, most of approaches learn the low-rank tensor representation and the final affinity matrix separately and ignore the difference between views. In this paper, we construct the target tensor composed of multiple normalized similarity matrices based on the Gaussian kernel function, which is constrained with the t-SVD based tensor nuclear norm to recover the low-rank part. The final affinity matrix is simultaneously learned via weighted multi-view fusion while optimizing the low-rank tensor, which suggests that each view is distributed to an adaptive weight. Moreover, the proposed method can be extended to semi-supervised classification through the collaborative optimization of the similarity tensor and the label indicator matrix. Extensive experiments conducted on four real-world datasets demonstrate the superiority of the proposed method compared with other state-of-the-art methods.

Social image colocalization locates the objects belonging to the same category from a set of images. Its goal is to explore the consistent relationship among social entities in the social system, which is conducive to promoting the development of social scene understanding. In a social relationship, consistency is easy to be ignored but is useful for many fields, e.g., natural language processing and computer vision. The social image colocalization aims to utilize the consistent relationship to discover the objects belonging to the same category and locate them by rectangle bounding boxes. The consistency is principally reflected in the category of objects, which can be achieved by a similar appearance or a uniform structure. Currently, state-of-the-art colocalization methods mainly focus on three solving strategies: candidate region proposal selection, saliency maps-based methods, and deep descriptor transformation. In this article, we sort out several kinds of the present mainstream methods and provide a comprehensive review of their fundamentals and performance. We expect that the overview of colocalization would be helpful for the researchers who are new to this area and provide them enlightenments.

Recently, 3D deep learning technologies require a large amount of supervised 3D point-cloud data to learn statistical models for various ITS-related tasks, e.g. object classification, object detection, object segmentation, etc. However, manually annotating 3D point-cloud data is time-consuming and labor-intensive. Therefore, this paper aims at co-locating 3D objects from mobile LiDAR point clouds without any help of supervised training data. To realize it, we propose a new framework to implement 3D object co-localization for automatically extracting the objects of the same category from different point-cloud scenes. Specifically, to search and exploit the co-information from objects in different point-cloud scenes, we formulate a 3D object co-localization problem as a maximal subgraph matching problem. During the graph construction procedure, to handle the inconsistent representation of objects in different scenes, we propose a multi-scale clustering method to represent objects by a pyramid structure. In addition, because the maximal subgraph matching problem is NP-hard, we propose a stochastic search algorithm to generate the co-localization results. Extensive experiments on the point-cloud data collected by the Reigl VMX450 mobile LiDAR system demonstrate the promising performance of the proposed framework.

Distance metric learning aims to learn a positive semidefinite matrix such that similar samples are preserved with small distances while dissimilar ones are mapped with big values above a predefined margin. It can facilitate to improve the performance of certain learning tasks. In this article, distance metric learning and clustering are integrated into an unified framework via rank-reduced regression. First, distance metric learning is proved to be consistent with rank-reduced regression, which provides a new perspective to learn structured regularization matrices. Second, orthogonal and non-negative rank-reduced regression problems are addressed individually for clustering, and the corresponding algorithms with proved convergence are proposed. Finally, both distance metric learning and clustering are addressed simultaneously in the problem formulation, which may trigger some new insights for learning an effective clustering oriented low-dimensional embedding. To show the superior performance of the proposed method, we compare it with several state-of-the-art clustering approaches. And, extensive experiments on the test datasets demonstrate the superiority of the proposed method.

Geometric model fitting has been widely used in many computer vision tasks. However, it remains as a challenging task when handing multiple-structural data contaminated by noises and outliers. Most previous work on model fitting cannot guarantee the consistency of their solutions due to their randomness, precluding them from many real-world applications. In this research, we propose a fast two-view approximately deterministic model fitting scheme (called LGF), to provide consistent solutions for multiple-structural data. The proposed LGF scheme starts from defining preference function by preserving local neighborhood relationship, and then adopts the min-hash technique to roughly sample subsets. By this way, it is able to cover all model instances in data in the parameter space with a high probability. After that, LGF refines the previous sampled subsets by globalresidual optimization. Furthermore, we propose a simple yet effective model selection framework to estimate the number and the parameters of model instances in data. Extensive experiments on real images show that the proposed LGF scheme is able to observe superior or very competitive performance on both accuracy and speed over several state-of-the-art model fitting methods.

High-dimensional features are extensively accessible in machine learning and computer vision areas. How to learn an efficient feature representation for specific learning tasks is invariably a crucial issue. Due to the absence of class label information, unsupervised feature representation is exceedingly challenging. In the last decade, deep learning has captured growing attention from researchers in a broad range of areas. Most of the deep learning methods are supervised, which is required to be fed with a large amount of accurately labeled data points. Nevertheless, acquiring sufficient accurately labeled data is unaffordable in numerous real-world applications, which is suggestive of the needs of unsupervised learning. Toward this end, quite a few unsupervised feature representation approaches based on deep learning have been proposed in recent years. In this paper, we attempt to provide a comprehensive overview of unsupervised deep learning methods and compare their performances in text categorization. Our survey starts with the autoencoder and its representative variants, including sparse autoencoder, stacked autoencoder, contractive autoencoder, denoising autoencoder, variational autoencoder, graph autoencoder, convolutional autoencoder, adversarial autoencoder, and residual autoencoder. Aside from autoencoders, deconvolutional networks, restricted Boltzmann machines, and deep belief nets are introduced. Then, the reviewed unsupervised feature representation methods are compared in terms of text clustering. Extensive experiments in eight publicly available data sets of text documents are conducted to provide a fair test bed for the compared methods.

High dimensionality is quite commonly encountered in data mining problems, and hence dimensionality reduction becomes an important task in order to improve the efficiency of learning algorithms. As a widely used technique of dimensionality reduction, feature selection is about selecting a feature subset being guided by certain criterion. In this paper, three unsupervised feature selection algorithms are proposed and addressed from the viewpoint of sparse graph embedding learning. First, using the self-characterization of the given data, we view the data themselves as a dictionary, conduct sparse coding and propose the sparsity preserving feature selection (SPFS) algorithm. Second, considering the locality preservation of neighborhoods for the data, we study a special case of the SPFS problem, namely, neighborhood preserving feature selection problem, and come up with a suitable algorithm. Third, we incorporate sparse coding and feature selection into one unified framework, and propose a neighborhood embedding feature selection (NEFS) criterion. Drawing support from nonnegative matrix factorization, the corresponding algorithm for NEFS is presented and its convergence is proved. Finally, the three proposed algorithms are validated with the use of eight publicly available real-world datasets from machine learning repository. Extensive experimental results demonstrate the superiority of the proposed algorithms over four compared state-of-the-art unsupervised feature selection methods.

Data fusion is used to integrate features from heterogeneous data sources into a consistent and accurate representation for certain learning tasks. As an effective technique for data fusion, unsupervised multimodal feature representation aims to learn discriminative features, indicating the improvement of classification and clustering performance of learning algorithms. However, it is a challenging issue since varying modality favors different structural learning. In this paper, we propose an efficient feature learning method to represent multimodal images as a sparse multigraph structure embedding problem. First, an effective algorithm is proposed to learn a sparse multigraph construction from multimodal data, where each modality corresponds to one regularized graph structure. Second, incorporating the learned multigraph structure, the feature learning problem for multimodal images is formulated as a form of matrix factorization. An efficient corresponding algorithm is developed to optimize the problem and its convergence is also proved. Finally, the proposed method is compared with several state-of-the-art single-modal and multimodal feature learning techniques in eight publicly available face image datasets. Comprehensive experimental results demonstrate that the proposed method outperforms the existing ones in terms of clustering performance for all tested datasets.

Rough sets are efficient to extract rules from information systems. Matroids generalize the linear independency in vector spaces and the cycle in graphs. Specifically, matroids provide well-established platforms for greedy algorithms, while most existing algorithms for many rough set problems including attribute reduction are greedy ones. Therefore, the combination between rough sets and matroids may bring new efficient solutions to those important and difficult problems. In this paper, 2-circuit matroids, abstracted from matroidal characteristics of rough sets, are studied and axiomatized. A matroid is induced by an equivalence relation, and its characteristics including the independent set and duality are represented with rough sets. Based on these rough set representations, this special type of matroid is defined as 2-circuit matroids. Conversely, an equivalence relation is induced by a matroid, and its relationship with the above induction is further investigated. Finally, a number of axioms of the 2-circuit matroid are obtained through rough sets. These interesting and diverse axioms demonstrate the potential for the connection between rough sets and matroids.