Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks486 from public data
- Deep Learning for Generic Object Detection: A Survey2,890
A comprehensive survey of the recent achievements in this field brought about by deep learning techniques, covering many aspects of generic object detection: detection frameworks, object feature representation, object proposal generation, context modeling, training strategies, and evaluation metrics.
- Improved Deep Embedded Clustering with Local Structure Preservation1,085
The Improved Deep Embedded Clustering (IDEC) algorithm can jointly optimize cluster labels assignment and learn features that are suitable for clustering with local structure preservation by integrating the clustering loss and autoencoder’s reconstruction loss.
- Deep Clustering with Convolutional Autoencoders543
A convolutional autoencoders structure is developed to learn embedded features in an end-to-end way and a clustering oriented loss is directly built on embedded features to jointly perform feature refinement and cluster assignment.
- 520
- 493
- Late Fusion Incomplete Multi-View Clustering374
This work proposes Late Fusion Incomplete Multi-view Clustering (LF-IMVC) which effectively and efficiently integrates the incomplete clustering matrices generated by incomplete views and develops a three-step iterative algorithm to solve the resultant optimization problem with linear computational complexity and theoretically prove its convergence.
- Scalable Multi-view Subspace Clustering with Unified Anchors362
This work proposes a Scalable Multi-view subspace Clustering with Unified Anchors (SMVSC), which combines anchor learning and graph construction into a unified optimization framework and achieves comparable or better clustering performance much more efficiently.
- A Survey of Knowledge Graph Reasoning on Graph Types: Static, Dynamic, and Multi-Modal354
A first survey for knowledge graph reasoning tracing from static to temporal and then to multi-modal KGs, and points out the challenges and potential opportunities to enlighten the readers.
- 339
- Deep Graph Clustering via Dual Correlation Reduction338
This work proposes a novel self-supervised deep graph clustering method termed Dual Correlation Reduction Network (DCRN) by reducing information correlation in a dual manner and introduces a propagation regularization term to enable the network to gain long-distance information with the shallow network structure.
- 319
- Deep Fusion Clustering Network307
A Deep Fusion Clustering Network (DFCN) is proposed, in which an interdependency learning-based Structure and Attribute Information Fusion (SAIF) module is proposed to explicitly merge the representations learned by an autoencoder and a graph autoen coder for consensus representation learning.
- Multiple Kernel k-means with Incomplete Kernels306
Two simple yet effective algorithms that perform multiple kernel clustering directly with the presence of incomplete kernel matrices, which are treated as auxiliary variables to be jointly optimized and which consistently achieve superior performance are proposed.
- 295
- 289
- Efficient One-Pass Multi-View Subspace Clustering with Consensus Anchors269
This work proposes a scalable and parameter-free MVSC method to directly output the clustering labels with optimal anchor graph, termed as Efficient One-pass Multi-view Subspace Clustering with Consensus Anchors (EOMSC-CA), which combines anchor learning and graph construction into a uniform framework to boost clustering performance.
- Global and Local Structure Preservation for Feature Selection263
This paper proposes a global and local structure preservation framework for feature selection (GLSPFS) which integrates both global pairwise sample similarity and local geometric data structure to conduct feature selection and shows that the best feature selection performance is always obtained when the two factors are appropriately integrated.
- Multiple Kernel k-Means Clustering with Matrix-Induced Regularization260
This paper proposes an MKKM clustering with a novel, effective matrix-induced regularization to reduce such redundancy and enhance the diversity of the selected kernels and shows that maximizing the kernel alignment for clustering can be viewed as a special case of this approach.
- Multi-view Clustering via Late Fusion Alignment Maximization257
This paper theoretically uncovers the connection between existing k-means clustering and the alignment between base partitions and consensus partition and proposes a simple but effective multi-view algorithm termed MVC-LFA, which proposes to maximally align the consensus partition with the weighted base partitions.
- CGD: Multi-View Clustering via Cross-View Graph Diffusion231
The proposed CGD takes the traditional predefined graph matrices of different views as input, and learns an improved graph for each single view via an iterative cross diffusion process by capturing the underlying manifold geometry structure of original data points, and leveraging the complementary information among multiple graphs.
- 225
- 218
- Simple Contrastive Graph Clustering209
A simple contrastive graph clustering algorithm to improve the existing methods from the perspectives of network architecture, data augmentation, and objective function, and a novel cross-view structural consistency objective function is designed to enhance the discriminative capability of the learned network.
- Multiple kernel extreme learning machine204
A general learning framework, termed multiple kernel extreme learning machines (MK-ELM), to address the lack of a general framework for ELM to integrate multiple heterogeneous data sources for classification and can achieve comparable or even better classification performance than state-of-the-art MKL algorithms, while incurring much less computational cost.
- 198
- Unsupervised feature selection via multiple graph fusion and feature weight learning191
This paper introduces a novel and effective unsupervised feature selection method via multiple graph fusion and feature weight learning (MGF^2WL) to address these issues, and develops a graph fusion term to fuse multiple predefined similarity graphs for learning a unified similarity graph.
- DealMVC: Dual Contrastive Calibration for Multi-view Clustering189
A novel Dual contrastive calibration network for Multi-View Clustering (DealMVC) is proposed, which first design a fusion mechanism to obtain a global cross-view feature, and proposes a global Contrastive calibration loss to constrain the consistency of pair-wise view features.
- Knowledge Graph Contrastive Learning Based on Relation-Symmetrical Structure166
A knowledge graph contrastive learning framework based on relation-symmetrical structure, KGE-SymCL, which mines symmetrical structure information in KGs to enhance the discriminative ability of KGE models and could outperform other state-of-the-art baselines.
- 165
- Structured graph learning for clustering and semi-supervised classification162
This paper proposes a graph learning framework to preserve both the local and global structure of data that uses the self-expressiveness of samples to capture the global structure and adaptive neighbor approach to respect the local structure.
- 162
- 154
- 154
- 150
- Align then Fusion: Generalized Large-scale Multi-view Clustering with Anchor Matching Correspondences149
This work proposes the first study of the generalized and flexible anchor graph fusion framework termed FMVACC, and shows how to find anchor correspondence with both feature and structure information, after which anchor graphs obtained from different views are not aligned column-wisely.
- 147
- Learn from Relational Correlations and Periodic Events for Temporal Knowledge Graph Reasoning145
A novel reasoning model, termed RPC, is proposed, which sufficiently mines the information underlying the Relational correlations and Periodic patterns via two novel Correspondence units, i.e., relational correspondence unit (RCU) and periodic correspondence units (PCU).
- Deep Incomplete Multi-View Clustering with Cross-View Partial Sample and Prototype Alignment144
A Cross-view Partial Sample and Prototype Alignment Network (CPSPAN) for Deep Incomplete Multi-view Clustering and a prototype alignment module to achieve incomplete distribution calibration across views are proposed.
- 144
- Unsupervised feature selection via latent representation learning and manifold regularization144
A robust unsupervised feature selection method which embeds the latent representation learning into feature selection and is carried out in the learned latent representation space which is more robust to noises.
- One-Step Multi-View Clustering With Diverse Representation141
This work proposes a one-step multi-view clustering with diverse representation (OMVCDR) method, which incorporates multi-view learning and k-means into a unified framework and develops an efficient optimization algorithm with proven convergence to solve the resultant problem.
- 141
- Robust unsupervised feature selection via dual self-representation and manifold regularization138
Experimental results on ten real-world data sets demonstrate that the proposed method can effectively identify important features, outperforming many state-of-the-art unsupervised feature selection methods in terms of clustering accuracy (ACC) and normalized mutual information (NMI).
- 133
- Optimal Neighborhood Kernel Clustering with Multiple Kernels133
An optimal neighborhood kernel clustering (ONKC) algorithm is proposed to enhance the representability of the optimal kernel and strengthen the negotiation between kernel learning and clustering and significantly outperforms the state-of-the-art ones in the literature.
- A Survey of Graph Neural Networks in Real World: Imbalance, Noise, Privacy and OOD Challenges132
This paper presents a comprehensive survey that systematically reviews existing GNN models, focusing on solutions to the four mentioned real-world challenges including imbalance, noise, privacy, and OOD in practical scenarios that many existing reviews have not considered.
- Improved autoencoder for unsupervised anomaly detection130
Different from previous methods, by integrating the anomaly detection‐based loss and autoencoder's reconstruction loss, IAEAD can jointly optimize for anomaly detection tasks and learn representations that preserve the local data structure to avoid feature distortion.
- 128
- 124
- 120
- Effective End-to-end Unsupervised Outlier Detection via Inlier Priority of Discriminative Network120
E^3Outlier for the first time leverages a discriminative DNN for better representation learning, by using surrogate supervision to create multiple pseudo classes from original unlabelled data, and proposes the negative entropy based score as a simple and effective outlierness measure.
- Deep Embedded Clustering with Data Augmentation118
This paper proposes the framework of Deep Embedded Clustering with Data Augmentation (DEC-DA), which first train an autoencoder with the augmented data to construct the initial feature space and constrain the embedded features with a clustering loss to further learn clustering-oriented features.
- 117
- Self-Supervised Temporal Graph Learning With Temporal and Structural Intensity Alignment116
This work proposes a self-supervised method called S2T for temporal graph learning, which extracts both temporal and structural information to learn more informative node representations and achieves at most 10.13% performance improvement compared with the state-of-the-art competitors on several datasets.
- Latent Semantic Aware Multi-View Multi-Label Classification115
A novel approach for multi-view multi-label learning based on matrix factorization to exploit complementarity among different views and provides theoretical guarantee for the strict convexity for this method by properly setting parameters.
- 110
- Self-Representation Subspace Clustering for Incomplete Multi-view Data109
It is widely observed that self-representation subspace method enjoys a better clustering performance over the factorization based one, so this work adapts it to incomplete data by jointly performing data imputation and self- Representation learning.
- Consensus learning guided multi-view unsupervised feature selection107
A consensus learning guided multi-view unsupervised feature selection method, which embeds multi-View feature selection into a non-negative matrix factorization based clustering with sparse constrain.
- Multi-view Clustering via Deep Matrix Factorization and Partition Alignment106
This work proposes a novel multi-view clustering algorithm via deep matrix decomposition and partition alignment, and an alternating optimization algorithm is developed to solve the optimization problem with proven convergence.
- 102
- 101
- Let the Data Choose: Flexible and Diverse Anchor Graph Fusion for Scalable Multi-View Clustering101
A novel anchor-based method termed Flexible and Diverse Anchor Graph Fusion for Scalable Multi-view Clustering (FDAGF) and a novel hybrid fusion strategy for multi-size anchor graphs with theoretical proof, which allows flexible and diverse anchor graph fusion.
- A Self-Supervised Gait Encoding Approach With Locality-Awareness for 3D Skeleton Based Person Re-Identification101
A self-supervised gait encoding approach that can leverage unlabeled skeleton data to learn gait representations for person re-identification via gait features within 3D skeleton sequences and a locality-aware attention mechanism and contrastive learning scheme, which aim to preserve locality-awareness on intra-sequence level and inter- sequence level respectively during self- supervised learning.
- 100
- Attributed Graph Clustering with Dual Redundancy Reduction97
A novel method termed attributed graph clustering with dual redundancy reduction (AGC-DRR) to reduce the information redundancy in both input space and latent feature space and an adversarial learning mechanism to adaptively learn a redundant edge-dropping matrix to ensure the diversity of the compared sample pairs.
- Fast Continual Multi-View Clustering With Incomplete Views93
Comprehensive experiments conducted on various datasets demonstrate the superiority of FCMVC-IV over the competing approaches, which maintains a scalable consensus coefficient matrix and updates its knowledge with the incoming incomplete view rather than storing and recomputing all the data matrices.
- CONVERT: Contrastive Graph Clustering with Reliable Augmentation92
A novel CONtrastiVe Graph ClustEring network with Reliable AugmenTation (COVERT) that distills reliable semantic information by recovering the perturbed latent embeddings and a label-matching mechanism is designed to guide the model by clustering information through aligning the semantic labels and the selected high-confidence clustering pseudo labels.
- 92
- 91
- Efficient and Effective Incomplete Multi-View Clustering89
The proposed Efficient and Effective Incomplete Multi-view Clustering (EE-IMVC) algorithm proposes to impute each incomplete base matrix generated by incomplete views with a learned consensus clustering matrix to solve the resultant optimization problem with linear computational complexity and theoretically prove its convergence.
- 88
- Multi-View Spectral Clustering with High-Order Optimal Neighborhood Laplacian Matrix88
This paper designs an efficient algorithm with proved convergence to solve the resultant optimization problem of the optimal Laplacian matrix, and extends it to the late fusion version for accurate and efficient multi-view clustering.
- Dink-Net: Neural Clustering on Large Graphs87
A scalable deep graph clustering method (Dink-Net) is proposed with the idea of dilation and shrink, unifying the two-step clustering, i.e., representation learning and clustering optimization, into an end-to-end framework, guiding the network to learn clustering-friendly features.
- Robust graph regularized unsupervised feature selection85
This work proposes a unified UFS framework via feature self-representation and robust graph regularization, with the aim at reducing the sensitivity to outliers from the following two aspects: an l2, 1-norm is used to characterize the feature representation residual matrix and an l1-norm based graph Laplacian regularization term is adopted to preserve the local geometric structure of data.
- 84
- 83
- 82
- 82
- 82
- scDFC: A deep fusion clustering method for single-cell RNA-seq data81
This work proposes a novel single-cell deep fusion clustering model, which contains two modules, i.e. an attributed feature clustering module and a structure-attention feature clustered module, and two elegantly designed autoencoders are built to handle both features regardless of their data types.
- Unpaired Multi-View Graph Clustering With Cross-View Structure Matching81
Unlike the existing methods, UPMGC-SM effectively utilizes the structural information from each view to refine cross-view correspondences, and is a unified framework for both the fully and partially unpaired multi-view graph clustering.
- 81
- K-Means Clustering With Incomplete Data81
A novel K-means based clustering algorithm is proposed which unifies the clustering and imputation into one single objective function and makes these two processes be negotiable with each other to achieve optimality.
- 79
- 79
- Deep Temporal Graph Clustering78
A general framework for deep Temporal Graph Clustering called TGC is proposed, which introduces deep clustering techniques to suit the interaction sequence-based batch-processing pattern of temporal graphs and shows that temporal graph clustering enables more flexibility in finding a balance between time and space requirements.
- Multiple kernel clustering with local kernel alignment maximization78
A novel MKC algorithm with a "local" kernel alignment, which only requires that the similarity of a sample to its k-nearest neighbours be aligned with the ideal similarity matrix, which helps the clustering algorithm to focus on closer sample pairs that shall stay together and avoids involving unreliable similarity evaluation for farther sample pairs.
- Cross-View Local Structure Preserved Diversity and Consensus Learning for Multi-View Unsupervised Feature Selection75
This paper proposes a cross-view local structure preserved diversity and consensus semantic learning model for MV-UFS, termed CRV-DCL, and carefully design an efficient algorithm with convergence guarantee to solve the resultant optimization problem.
- 73
- 72
- 71
- 71
- Local Sample-Weighted Multiple Kernel Clustering With Consensus Discriminative Graph70
A novel local sample-weighted MKC (LSWMKC) model is proposed that possesses better local manifold representation and outperforms existing kernel or graph-based clustering algorithms.
- Deep Embedded Complementary and Interactive Information for Multi-View Classification70
This work proposes a novel multi-view learning framework that seamlessly embeds various view-specific information and deep interactive information and introduces a novelmulti-view fusion strategy to make a joint decision during the optimization for classification.
- Dual Test-Time Training for Out-of-Distribution Recommender System68
This paper proposes a novel Dual Test-Time-Training framework for OOD Recommendation, termed DT3OR, which is the first work to address OOD recommendation via a test-time-training strategy and conducts experiments on five datasets with various backbones.
- Continual Multi-view Clustering68
This work proposes a continual approach on the basis of late fusion multi-view clustering framework that only needs to maintain a consensus partition matrix and update knowledge with the incoming one of a new data view rather than keep all of them.
- 68
- 67
- 67
- 66
- A Survey of Deep Graph Clustering: Taxonomy, Challenge, Application, and Open Resource66
The detailed definition of deep graph clustering and the important baseline methods are introduced and the taxonomy of deepgraph clustering methods is proposed based on four different criteria including graph type, network architecture, learning paradigm, and clustering method.
- Interpolation-Based Contrastive Learning for Few-Label Semi-Supervised Learning66
An interpolation-based method to construct more reliable positive sample pairs and a novel contrastive loss to guide the embedding of the learned network to change linearly between samples so as to improve the discriminative capability of the network by enlarging the margin decision boundaries are proposed.
- 66
- 65
- 65
- DeepKSPD: Learning Kernel-Matrix-Based SPD Representation For Fine-Grained Image Recognition65
A deep network that jointly learns local descriptors, kernel-matrix-based SPD representation, and the classifier via an end-to-end training process is proposed and the derivatives for the mapping from a local descriptor set to the SPD representation to carry out backpropagation are derived.
- A Survey of Deep Graph Clustering: Taxonomy, Challenge, Application, and Open Resource64
A comprehensive survey of deep graph clustering methods, introducing formulaic definition, evaluation, and development in this field, and providing open resource supports, including a collection of state-of-the-art DGC methods and a flexible and extensible Python library for DGC.
- 64
- 64
- Multi-view subspace clustering via adaptive graph learning and late fusion alignment63
This work proposes a multi-view subspace clustering method via adaptive graph learning and late fusion alignment (AGLLFA), where for each view, AGLLFA learns an affinity graph adaptively to capture the similarity relationship among samples.
- Multi-View Spectral Clustering with Optimal Neighborhood Laplacian Matrix63
This paper proposes a novel optimal neighborhood multi-view spectral clustering algorithm, which enhances the representative capacity of the optimal Laplacian and better utilizes the hidden high-order connection information, leading to improved clustering performance.
- Subspace segmentation-based robust multiple kernel clustering62
This paper unify the constrained kernel polarization and subspace segmentation into a single procedure, where the resultant affinity matrix embedded with robust subspace structural information is utilized to guide the linear combination of base kernels.
- Attribute-Missing Graph Clustering Network60
This work proposes a novel Attribute-Missing Graph Clustering (AMGC) method to alternately promote clustering and imputation in a unified framework, where it iteratively produces the clustering-enhanced nearest neighbor information to conduct the data imputation process and utilizes the imputed information to implicitly refine the clustering distribution through model optimization.
- 60
- Simultaneous Representation Learning and Clustering for Incomplete Multi-view Data59
A Simultaneous Representation Learning and Clustering method that learns lowdimensional representations of present instances on each view and a common probability label matrix simultaneously, so that the nonlinear information can be reflected by these representations and the clustering results can obtained from label matrix directly.
- Sample-Level Cross-View Similarity Learning for Incomplete Multi-View Clustering58
A Sample-level Cross-view Similarity Learning (SCSL) method for Incomplete Multi-view Clustering that simultaneously construct a complete similarity matrix across views based on the inter-view sample relationship and the intra-view sample relationship and a simultaneously learning consensus representation ensures the validity of the projection.
- Localized Incomplete Multiple Kernel k-means57
Different from existing MKKM-IK, LI-MKKM only requires the similarity of a sample to its k-nearest neighbors to align with their ideal similarity values, which helps the clustering algorithm to focus on closer sample pairs that shall stay together and avoids involving unreliable similarity evaluation for farther sample pairs.
- End-to-end Learnable Clustering for Intent Learning in Recommendation56
This work proposes a novel intent learning method termed ELCRec, which improves NDCG@5 by 8.9\% and reduces computational costs by 22.5\% on the Beauty dataset and deploys this method on the industrial recommendation system with 130 million page views and achieves promising results.
- Efficient Multi-View Graph Clustering with Local and Global Structure Preservation56
A unified framework with a theoretical guarantee is designed to capture local and global information in anchor-based multi-view graph clustering, which inherits the linear complexity of existing AMVGC methods respecting the sample number and scales well with the data size.
- 55
- One-Stage Incomplete Multi-view Clustering via Late Fusion54
This paper proposes to unify the imputation of incomplete views and the clustering task into a single optimization procedure, so that the learning of the consensus partition matrix can directly assist the final clusteringtask.
- Scalable Incomplete Multi-View Clustering with Structure Alignment53
A novel incomplete anchor graph learning framework termed Scalable Incomplete Multi-View Clustering with Structure Alignment (SIMVC-SA), which constructs the view-specific anchor graph to capture the complementary information from different views and proposes a novel structure alignment module to refine the cross-view anchor correspondence.
- Initializing Then Refining: A Simple Graph Attribute Imputation Network53
A novel graph-oriented imputation framework called initializing then refining (ITR), where the structure information is employed for initial imputation, and the affinity structure is adaptively updated through iterations by calculating the sample-wise correlations upon the recomposed embeddings.
- 52
- Reinforcement Graph Clustering with Unknown Cluster Number51
This work proposes a new deep graph clustering method termed Reinforcement Graph Clustering (RGC), where cluster number determination and unsupervised representation learning are unified into a uniform framework by the reinforcement learning mechanism.
- Deep Anomaly Discovery from Unlabeled Videos via Normality Advantage and Self-Paced Refinement51
A full deep neural network (DNN) based solution that can realize highly effective UVAD and a novel self-paced refinement (SPR) scheme, which is synthesized into LBR to conduct UVAD, which enhances the UVAD performance and sometimes even surpasses the best classic VAD methods.
- Hierarchical Multiple Kernel Clustering51
This paper gradually group samples into fewer clusters, together with generating a sequence of intermediary matrices of descending sizes, and proposes a hierarchical approach to perform clustering while preserving advantageous details maximumly.
- 50
- 50
- Object Detection in Hyperspectral Image via Unified Spectral–Spatial Feature Aggregation49
This study presents a novel spectral–spatial aggregation (S2ADet) object detector that effectively harnesses the rich spectral and spatial complementary information inherent in the HSI, and proposes a feature aggregation two-stream network for interacting spectral-spatial features.
- FISH-MML: Fisher-HSIC Multi-View Metric Learning49
This work presents a simple yet effective model for multi-view metric learning, which aims to improve the classification of data with multiple views, e.g., multiple modalities or multiple types of features, based on Fisher discriminant analysis and Hilbert-Schmidt Independence Criteria.
- Structure-Adaptive Multi-View Graph Clustering for Remote Sensing Data48
A Structure-Adaptive Multi-View Graph Clustering method named SAMVGC on remote sensing data which boosts the structure homogeneity and cluster compactness by adaptively learning the graph and cluster structures, respectively.
- TMac: Temporal Multi-Modal Graph Learning for Acoustic Event Classification47
This work proposes a Temporal Multi-modal graph learning method for Acoustic event Classification, called TMac, by modeling such temporal information via graph learning techniques, and constructs a temporal graph for each acoustic event, dividing its audio and video data into multiple segments.
- Graph regularized spatial–spectral subspace clustering for hyperspectral band selection47
A graph regularized spatial-spectral subspace clustering method for hyperspectral band selection, which adopts superpixel segmentation to preserve the spatial information of HSIs by segmenting their first principal component into diverse homogeneous regions, and which can mitigate the effect of noise on the band selection.
- MINES: Message Intercommunication for Inductive Relation Reasoning over Neighbor-Enhanced Subgraphs46
This work proposes a novel GraIL-based framework, termed MINES, by introducing a Message Intercommunication mechanism on the Neighbor-Enhanced Subgraph to capture the omitted hidden mutual information and introduces bi-directed information interactions between connected entities.
- Mixed Graph Contrastive Network for Semi-supervised Node Classification45
This work proposes a novel graph contrastive learning method, termed Mixed Graph Contrastive Network (MGCN), which improves the discriminative capability of the latent embeddings by an interpolation-based augmentation strategy and a correlation reduction mechanism.
- 45
- Robust Graph-Based Multi-View Clustering45
This paper defines a min-max formulation for robust learning and rewrite it as a convex and differentiable objective function whose convexity and differentiability are carefully proved and can efficiently solve the resultant problem using a reduced gradient descent algorithm, and the corresponding solution is guaranteed to be globally optimal.
- Reliable Multi-View Clustering45
Experimental results on a number of data sets demonstrate that the proposed Reliable Multi-View Clustering method can effectively improve the reliability of multi-view clustering.
- 44
- 44
- Multi-view clustering with extreme learning machine43
This paper proposes a novel multi-view clustering framework with ELM to further improve clustering performance, and implements three algorithms based on this framework that yield superior clusteringperformance when compared with several state-of-art multi- view clustering methods in recent literatures.
- 42
- 42
- PR-ELM: Parallel regularized extreme learning machine based on cluster42
This paper proposes two parallel variants of ELM, termed as data parallel regularized ELM (DPR-ELM) and model parallel regularization ELm (MPR- ELM), to further improve the computational efficiency of ELm in handling large scale learning tasks.
- Metamaterial terahertz switch based on split-ring resonator embedded with photoconductive silicon42
A metamaterial terahertz switch based on a split-ring resonator embedded with photoconductive silicon is presented and numerically investigated and is frequency tunable by changing the thickness and permittivity of the dielectric layer.
- An Efficient and Robust Unsupervised Anomaly Detection Method Using Ensemble Random Projection in Surveillance Videos41
This paper proposes a novel efficient three-stage unsupervised anomaly detection method that embraces random projection instead of autoencoder or its variants in previous works and achieves comparable performance contrast with the state-of-the-art unsuper supervised method with much less running time.
- 40
- 39
- ARISE: Graph Anomaly Detection on Attributed Networks via Substructure Awareness39
This work proposes a new graph anomaly detection framework on attributed networks via substructure awareness (ARISE), which focuses on the substructures in the graph to discern abnormalities and introduces a graph contrastive learning scheme, which observes attribute anomalies in the meantime.
- Financial time series prediction using ℓ2,1RF-ELM39
The underlying rules of the correlation between cash inflow and outflow that can help improve accuracy, which is valuable for financial institutions to predict the trend of liquidity, are found.
- 38
- Game-Theoretic Design of Optimal Two-Sided Rating Protocols for Service Exchange Dilemma in Crowdsourcing38
This paper develops the first game-theoretic design of the two-sided rating protocol to stimulate cooperation among self-interested users, which consists of a recommended strategy and a rating update rule.
- Sample-adaptive Multiple Kernel Learning38
This work proposes a sample-adaptive MKL algorithm, in which base kernels are allowed to be adaptively switched on/off with respect to each sample, and which consistently outperforms the comparable ones in the literature.
- scEGG: an exogenous gene-guided clustering method for single-cell transcriptomic data37
This research introduces an innovative deep clustering method for single-cell data that leverages exogenous gene information to generate discriminative cell representations and provides enhanced insights into the characteristics of cells, thus laying the foundation for the early diagnosis and treatment of diseases.
- 37
- Deep Temporal Graph Clustering: A Comprehensive Benchmark and Datasets36
A BenchTGC Framework is designed to illustrate the paradigm of temporal graph clustering and improve existing clustering techniques to fit temporal graphs, and discusses problems with public temporal graph datasets.
- 36
- RARE: Robust Masked Graph Autoencoder36
A novel SGP method termed Robust mAsked gRaph autoEncoder (RARE) is proposed to improve the certainty in inferring masked data and the reliability of the self-supervision mechanism by further masking and reconstructing node samples in the high-order latent feature space.
- Hyperspectral Band Selection via Spatial-Spectral Weighted Region-wise Multiple Graph Fusion-Based Spectral Clustering36
A hyperspectral band selection method via spatial-spectral weighted region-wise multiple graph fusion-based spectral clustering, referred to as RMGF briefly, to validate the superiority of the proposed method when compared with other state-of-the-art ones.
- Multiple kernel learning with hybrid kernel alignment maximization36
This work defines the local kernel alignment based on centralized kernel alignment and proposes an alternative algorithm with proved convergence to identify the multiple kernel coefficients that is superior to those of existing MKL algorithms.
- Random Fourier extreme learning machine with ℓ2,1 -norm regularization36
This paper introduces the random Fourier mappings as activation functions to approximate the Gaussian kernel, and adopts the ?
- Self-paced Adaptive Bipartite Graph Learning for Consensus Clustering35
This article proposes a novel self-paced consensus clustering method with adaptive bipartite graph learning to gradually involve data from more reliable to less reliable ones in consensus learning.
- RiskQ: Risk-sensitive Multi-Agent Reinforcement Learning Value Factorization34
RiskQ is proposed, which models the joint return distribution by modeling quantiles of it as weighted quantile mixtures of per-agent return distribution utilities and satisfies the RIGM principle for the VaR and distorted risk metrics.
- 34
- 33
- GraphLearner: Graph Node Clustering with Fully Learnable Augmentation33
GraphLearner introduces learnable augmentors to generate high-quality and task-specific augmented samples for CDGC and introduces an adversarial learning mechanism within the method to refine the high-confidence matrices.
- Simple Yet Effective: Structure Guided Pre-trained Transformer for Multi-modal Knowledge Graph Reasoning33
SGMPT is the first MPT for multi-modal KGR, which mines structural information underlying MKGs, and outperforms existing state-of-the-art models, and proves the effectiveness of the designed strategies.
- DPNET: Dynamic Poly-attention Network for Trustworthy Multi-modal Classification33
A novel Dynamic Poly-attention Network (DPNET) is proposed for trustworthy multi-modal classification and achieves competitive performance compared to other state-of-the-art ones.
- Tensor-Based Multi-View Block-Diagonal Structure Diffusion for Clustering Incomplete Multi-View Data32
- 31
- DVSAI: Diverse View-Shared Anchors Based Incomplete Multi-View Clustering30
This work designs a novel framework called Diverse View-Shared Anchors based Incomplete multi-view clustering (DVSAI), and devise an integration scheme with linear computational and memory expenditures to integrate the outputted multi-scale unified anchor graphs such that running spectral algorithm generates the spectral embedding.
- Automatic and Aligned Anchor Learning Strategy for Multi-View Clustering30
This work introduces a Hierarchical Bipartite Neighbor Clustering (HBNC) strategy to adaptively select a suitable number of representative anchors in each view and proposes an innovative anchor alignment strategy in response to the varying quantities of anchors across different views.
- Revisiting Initializing Then Refining: An Incomplete and Missing Graph Imputation Network30
A novel graph imputation network termed revisiting initializing then refining (RITR), where both attribute-incomplete and attribute-missing samples are completed under the guidance of a novel initializing-then-refining imputation criterion, which is the first end-to-end unsupervised framework dedicated to handling hybrid-absent graphs.
- Normality Learning-based Graph Anomaly Detection via Multi-Scale Contrastive Learning30
This work proposes a normality learning-based GAD framework via multi-scale contrastive learning networks (NLGAD for abbreviation) that improves the detection performance and learns a more accurate estimate of normality so that anomalous nodes can be more easily distinguished.
- An Incremental Feature Learning Algorithm Based on Least Square Support Vector Machine30
An incremental feature learning algorithm based on Least Square Support Vector Machine is proposed, which is more suitable to deal with classification tasks which can not be well solved by using a single kernel function.
- MS-Former: Memory-Supported Transformer for Weakly Supervised Change Detection With Patch-Level Annotations29
A memory-supported transformer (MS-Former), a novel framework consisting of a bi-directional attention block (BAB) and a patch-level supervision scheme (PSS) tailored for weakly supervised CD with patch-level annotations, which improves the performance of fully supervised change detection methods.
- R²MRF: Defocus Blur Detection via Recurrently Refining Multi-Scale Residual Features29
A deep neural network which Recurrently Refines Multi-scale Residual Features (R2MRF) for defocus blur detection consistently and significantly outperforms the competitors in terms of both efficiency and accuracy.
- Dual graph regularized compact feature representation for unsupervised feature selection29
An efficient optimization algorithm based on Alternating Direction Method of Multipliers is developed to solve the proposed optimization problem and experiments on various types of real world datasets are conducted to demonstrate the effectiveness of the proposed method.
- Triangle Lasso for Simultaneous Clustering and Optimization in Graph Datasets29
This paper demonstrates through extensive numerical experiments that triangle lasso is robust to the imperfect data, and develops an efficient algorithm based on Alternating Direction Method of Multipliers (ADMM) to obtain a moderately accurate solution.
- Large Margin Subspace Learning for feature selection29
A novel feature selection algorithm, termed as Large Margin Subspace Learning (LMSL), which seeks a projection matrix to maximize the margin of a given sample, defined as the distance between the nearest missing and the nearest hit of the given sample.
- A Lightweight Deep Exclusion Unfolding Network for Single Image Reflection Removal28
A novel Deep Exclusion unfolding Network (DExNet), a lightweight, interpretable, and effective network architecture for SIRR, which achieves state-of-the-art visual and quantitative results while utilizing only approximately 8% of the parameters required by leading methods.
- 28
- 28
- Efficient Anchor Learning-based Multi-view Clustering -- A Late Fusion Method28
The proposed Efficient Anchor Learning-based Multi-view Clustering (AL-MVC) algorithm achieves the state-of-the-art performance in both clustering performance and efficiency.
- Anomaly Detection With Kernel Preserving Embedding28
A novel anomaly detection method is proposed, which exploits kernel preserving embedding, as well as the double nuclear norm, to explore the similarity relations of data, and can not only preserve the manifold structural properties of the data, but also alleviate the suboptimal problem.
- 28
- SARF: Aliasing Relation–Assisted Self-Supervised Learning for Few-Shot Relation Reasoning27
A novel Self-supervised learning model by leveraging Aliasing Relations to assist FS-KGR, termed SARF is proposed, and extensive experiments demonstrate that SARF achieves state-of-the-art (SOTA) performance compared with other methods in most cases.
- 27
- Semantic instance segmentation with discriminative deep supervision for medical images27
A proposal-free segmentation network with discriminative deep supervision (DDS), which at the same time allows us to gain the power of the proposal-based method, and robust pixel-wise instance-level cues are introduced for semantic segmentation.
- Outlier Detection Ensemble with Embedded Feature Selection27
Comprehensive experimental results on 12 real-world datasets from diverse domains validate the superiority of the proposed ODEFS, which unifies feature selection and outlier detection into a pairwise ranking formulation to learn feature subsets that are tailored for the outlier Detection method.
- Large-scale k-means clustering via variance reduction27
This paper proposes a position correction mechanism to correct the drift of the cluster centers, and proposes a variance reduced k-means named VRKM, which outperform the state-of-the-art method and obtain about 2 × and 4 × speedups for large-scale clustering, respectively.
- Local kernel alignment based multi-view clustering using extreme learning machine27
An ELM based multi-view learning approach with different views generated by ELM random feature mapping with respect to different hidden-layer nodes is proposed, and the effectiveness and superiority of maximizing local kernel alignment on those views constructed by the proposed method are verified.
- DaRec: A Disentangled Alignment Framework for Large Language Model and Recommender System26
This work proves that directly aligning the representations of LLMs and collaborative models is suboptimal for enhancing downstream recommendation tasks performance, based on the information theorem, and proposes a novel plug-and-play alignment framework for LLMs and collaborative models.
- 26
- 25
- 25
- Efficient Multiple Kernel k-Means Clustering With Late Fusion25
This work designs two multiple kernel k-means clustering algorithms with late fusion, whose computational complexities linearly grow with the number of samples, and integrally optimize the various clustering matrices into the optimal consensus clustering results iteratively.
- A Computational Model of the Short-Cut Rule for 2D Shape Decomposition25
A computational model for the short-cut rule is implemented and it is demonstrated that, compared with state-of-the-art shape decomposition methods, the proposed approach achieves decomposition results, which better correspond to human intuition as revealed in psychological experiments.
- 25
- Distilling Reasoning Ability From Large Language Models With Adaptive Thinking24
A plug-and-play adaptive-thinking mechanism is proposed to integrate the merits of pre-thinking and post-thinking, in which a perception module based on soft prompt tuning is introduced to prompt SLM to answer or think first according to the complexity of questions.
- 24
- Hawkes-Enhanced Spatial-Temporal Hypergraph Contrastive Learning Based on Criminal Correlations24
Hawkes-enhanced Spatial-Temporal Hypergraph Contrastive Learning framework (HCL), which mines the aforementioned correlations via two specific strategies via two specific strategies, and shows promising capacities from four aspects, i.e., superiority, transferability, effectiveness, and sensitivity.
- 24
- 23
- DisARM: Displacement Aware Relation Module for 3D Detection23
Displacement Aware Relation Module (DisARM), a novel neural network module for enhancing the performance of 3D object detection in point cloud scenes, finds that training with relations only between the most representative ones, or an-chors, can significantly boost the detection performance.
- 23
- Neighborhood Consensus Networks for Unsupervised Multi-view Outlier Detection23
This paper proposes an unsupervised multi-view outlier detection method based on the proposed novel neighborhood consensus networks termed NC-Nets, which automatically encodes intrinsic information into a comprehensive latent space for each view and uniforms the neighborhood structures among different views.
- Incremental multiple kernel extreme learning machine and its application in Robo-advisors23
An incremental multiple kernel extreme learning machine (IMK-ELM) model, which initializes a generic training database and then tunes itself to the classification task, and is able to efficiently deal with large-scale tasks like Robo-advisors.
- Contrastive Continual Multiview Clustering With Filtered Structural Fusion22
A data buffer to store filtered structural information is developed and utilized to guide the generation of a robust partition matrix via contrastive learning and a sampling strategy called clustering then sample is proposed to address the high complexity involved in acquiring and storing structural information.
- 22
- 22
- 22
- E3 Outlier: a Self-Supervised Framework for Unsupervised Deep Outlier Detection22
This paper proposes a novel framework named E 3 Outlier that can perform effective and end-to-end deep outlier removal and introduces self-supervision into deep OD, and argues that inlier priority, a property that encourages DNN to prioritize inliers duringSelf-supervised learning, makes it possible to perform end- to-end OD.
- Reliable Attribute-missing Multi-view Clustering with Instance-level and feature-level Cooperative Imputation21
The proposed RAM-MVC method outperforms nine baseline methods, as evidenced by real-world experiments using single-cell multi-view data and a reliable imputation guidance module that distinguishes between real and technical attribute-missing events, enabling discriminative imputation.
- 21
- 21
- Diversity and consistency learning guided spectral embedding for multi-view clustering21
This work proposes a novel multi-view clustering method (DCMSC) by constraining diversity and consistency in both the learned clustering label matrix and data space, which suppresses the effect of noises and decreases redundant information.
- Consensus Kernel K -Means Clustering for Incomplete Multiview Data21
A new unified learning method is proposed for incomplete multiview clustering, which simultaneously imputes the incomplete views and learns a consistent clustering result with explicit modeling of between-view consistency.
- Multiple Kernel Clustering Framework with Improved Kernels21
A simple while effective framework to adaptively improve the quality of base kernels and fills the gap between kernel adaption and clustering procedure for the first time in the literature and is readily extendable.
- Incorporation of radius-info can be simple with SimpleMKL21
This paper improves SimpleMKL to utilize the information of the radius of MEB in an efficient and practical way and incorporates its close relative, the trace of data scattering matrix, to avoid the above problems.
- Efficient Prompt Tuning of Large Vision-Language Model for Fine-Grained Ship Classification20
A novel prompt tuning technique is introduced that employs a hierarchical, multigranularity prompt design and enhances the model’s generalization capabilities while improving its ability to discern intricate backgrounds and learn discriminative ship features.
- 20
- Fusion Multiple Kernel K-means20
A novel algorithm is proposed, termed as Fusion Multiple Kernel k-means (FMKKM), which unifies base partition learning and late fusion clustering into one single objective function, and adopts early fusion technique to capture more sufficient information in kernel matrices.
- 20
- 19
- G$^2$uardFL: Safeguarding Federated Learning Against Backdoor Attacks through Attributed Client Graph Clustering19
G$^2$uardFL is presented, a protective framework that reinterprets the identification of malicious clients as an attributed graph clustering problem, thus safeguarding FL systems and demonstrating its significant effectiveness in mitigating backdoor attacks while having a negligible impact on the aggregated model's performance on benign samples.
- Structure Guided Multi-modal Pre-trained Transformer for Knowledge Graph Reasoning19
SGMPT is the first MPT model for multimodal KGR, which mines the structural information underlying the knowledge graph, and outperforms existing state-of-the-art models, and proves the effectiveness of the designed strategies.
- Adaptive partial graph learning and fusion for incomplete multi‐view clustering19
This study presents a novel incomplete multi‐view clustering method via adaptive partial graph learning and fusion (APGLF), which can capture the local data structure of both within‐view and cross‐view as well as recover the optimal cluster structure of original data.
- DMP-ELMs: Data and model parallel extreme learning machines for large-scale learning tasks19
The goal is to build scalable ELMs with a large number of samples and hidden neurons, parallel running on clusters without computational and memory bottlenecks while having the same output results as the sequential ELM.
- A subspace ensemble framework for classification with high dimensional missing data19
This paper proposes a complete-case projection subspace ensemble framework, where two alternative partition strategies are developed for incomplete datasets with even missing patterns and uneven missing patterns, respectively, and significantly outperforms existing imputation methods in most cases.
- Distance Based Multiple Kernel ELM: A Fast Multiple Kernel Learning Approach19
A distance based multiple kernel extreme learning machine (DBMK-ELM), which provides a two-stage multiple kernel learning approach with high efficiency and superior performance in terms of the accuracy and the computational cost.
- Single-Cell Multi-View Clustering via Community Detection With Unknown Number of Clusters18
The scUNC method comprises several steps: initially, it employs a cross-view fusion network to create an effective embedding, which is then utilized to generate initial clusters via community detection, and subsequently, the clusters are automatically merged and optimized until no further clusters can be merged.
- 18
- Learning Knowledge-diverse Experts for Long-tailed Graph Classification18
The Knowledge-diverse Experts (KDEX) for long-tailed graph classification introduces a knowledge-diverse expert training approach to train experts with different capacities in managing various test distributions and outperforms current baselines in both standard and test-agnostic long-tailed graph classification.
- Deep Embedding for Determining the Number of Clusters18
Deep Embedding Determination first combines the virtues of the convolutional autoencoder and the t-SNE technique to extract low dimensional embedded features and determines the number of clusters using an improved density-based clustering algorithm.
- Integrative Multi‐Omics and Routine Blood Analysis Using Deep Learning: Cost‐Effective Early Prediction of Chronic Disease Risks17
This work highlights the role of multiomics analysis in the prediction of chronic disease risk, and the development and validation of predictive models based on blood routine results can help advance personalized medicine and reduce the cost of disease screening in the community.
- 17
- 17
- 17
- Multiple kernel clustering with corrupted kernels17
This paper proposes a novel method to learn an optimal consensus kernel from a group of pre-specified kernel matrices, each of which can be decomposed into the optimal consensuskernel matrix and a sparse error matrix and proposed scheme to address the problem of considerable corrupted kernels.
- MI-ELM: Highly efficient multi-instance learning based on hierarchical extreme learning machine17
Experimental results over several real-world data sets demonstrate that the proposed MI-ELM method significantly outperforms existing kernel methods for MI classification in terms of the classification accuracy and the classification time.
- 16
- 16
- Simultaneous Image-to-Zero and Zero-to-Noise: Diffusion Models with Analytical Image Attenuation16
This paper presents a novel diffusion paradigm named DDM based on the Itô diffusion process, in which the image distribution is approximated by an explicit transition probability while the noise path is controlled by the standard Wiener process, and finds that decoupling the diffusion process reduces the learning difficulty and the explicit transition probability improves the generative speed significantly.
- Understand Dynamic Regret with Switching Cost for Online Decision Making16
A new theoretical analysis framework is provided that shows an interesting observation; that is, the relation between the switching cost and the dynamic regret is different for settings of OA and OCO.
- Joint Multi-view 2D Convolutional Neural Networks for 3D Object Classification16
This work proposes a novel multi-view framework by jointly using multiple 2D-CNNs to capture discriminative information with relationships as well as a new multi-View loss fusion strategy, in an end-to-end manner that is able to outperform current state-of-the-art methods on 3D object classification.
- Affine Equivariant Autoencoder16
An affine equivariant autoencoder is proposed to learn features that areEquivariant to the affine transformation in an unsupervised manner to fill the gap in learning transformation invariant features in deep neural networks.
- 16
- Multiple Kernel Clustering With Global and Local Structure Alignment16
A global and local structure alignment framework for MKC (GLSAMKC) is proposed which well considers both the alignment between the global structure andLocal structure of data with the same ideal similarity matrix.
- Grouping-based adaptive spatial formation of swarm robots in a dynamic environment16
This work proposes an adaptive mechanism with three basic behaviours for each individual robot and designs a grouping-based spatial formation algorithm for swarm robots to respond to changes and accomplish shape formation, demonstrating the effectiveness of the proposed adaptive mechanism.
- Absent Multiple Kernel Learning16
This paper proposes an absent MKL (AMKL) algorithm that directly classifies each sample with its observed channels and shows that this problem can be reformulated as a convex one by applying the representer theorem, which makes it readily be solved via existing convex optimization packages.
- 15
- 15
- Heterogeneous Graph Guided Contrastive Learning for Spatially Resolved Transcriptomics Data15
A heterogeneous Graph guided Contrastive Learning (stGCL) for aggregating spatial transcriptomics data is constructed, guided by the inherent heterogeneity of cellular molecules by dynamically coordinating triple-level node attributes through comparative learning loss distributed across view domains, thus maintaining view independence during the aggregation process.
- A Lightweight Anchor-Based Incremental Framework for Multi-view Clustering15
This work proposes A Lightweight Anchor-Based Incremental Framework for Multi-view Clustering, and designs a three-step iterative and convergent algorithm to address the resultant problem of anchor alignment.
- 14
- Max-Mahalanobis Anchors Guidance for Multi-View Clustering14
A novel method called Max-Mahalanobis Anchors Guidance for multi-view Clustering (MAGIC), which guides the cross-view representations to progressively align with well-defined anchors, significantly enhancing the performance of multi-view clustering.
- 14
- 14
- 14
- 14
- 14
- Message Intercommunication for Inductive Relation Reasoning14
This work proposes a novel GraIL-based inductive relation reasoning model, termed MINES, by introducing a Message Intercommunication mechanism on the Neighbor-Enhanced Subgraph, designed to capture the omitted hidden mutual information.
- GANN: Graph Alignment Neural Network for Semi-Supervised Learning14
The Graph Alignment Neural Network (GANN) is proposed, a simple and effective graph neural architecture that can achieve considerable benefits in semi-supervised node classification and outperform state-of-the-art competitors.
- 14
- Predicting Future Locations with Semantic Trajectories14
This work designs ST-LSTM, which draws on semantic trajectories to predict future locations, and proposes a strategic filling algorithm to solve the cold-start problem.
- 14
- Approximate Large-scale Multiple Kernel k-means Using Deep Neural Network14
An approximate algorithm is proposed that trains a deep neural network to regress the indicating matrix generated by MKC algorithms on a small subset, and obtains the approximate indicating matrix of the whole data set using the trained network, and finally performs the k-means on the output of the network.
- An efficient radius-incorporated MKL algorithm for Alzheimer׳s disease prediction14
An improved radius-incorporated MKL algorithm for AD prediction is proposed that can better utilise the radius information and achieve higher prediction accuracy than the comparable MKL methods in the literature and demonstrates the highest computational efficiency among all the comparable methods.
- 13
- 13
- Alleviate Anchor-Shift: Explore Blind Spots with Cross-View Reconstruction for Incomplete Multi-View Clustering13
This work proposes a cross-view reconstruction strategy that not only alleviate the anchor shift problem through a carefully designed cross-view learning process, but also reconstructs missing samples in a way that transcends the limitations imposed by convex combinations.
- Sample Weighted Multiple Kernel K-means via Min-Max optimization13
A novel sample weighted multiple kernel k-means via min-max optimization (SWMKKM), which sufficiently considers the sum of relationship between one sample and the others to represent the sample weights, helps clustering algorithm pay more attention to samples with more positive effects on clustering and avoids unreliable overestimation for samples with poor quality.
- Partial multiview clustering with locality graph regularization13
An effective iterative algorithm is proposed to solve the resultant optimization problem with theoretically guaranteed convergence to address urgent issues of incomplete multiview clustering and locality graph regularization.
- 13
- A novel multi-view clustering method via low-rank and matrix-induced regularization13
A novel multi-view clustering method that construct a low-rank consensus matrix and a sparse error matrix from each similarity matrix corresponding to each view to reduce the redundancy and enhance the diversity among different views is proposed.
- Sample-Based Extreme Learning Machine with Missing Data13
A sample-based learning framework is proposed and two sample- based ELM algorithms for classification and regression are developed, which do more accurate and stable learning than other state-of-the-art ones, especially in the case of higher missing ratio.
- 13
- 12
- View Gap Matters: Cross-view Topology and Information Decoupling for Multi-view Clustering12
A Tree-Based View-Gap Maintaining Multi-View Clustering (TGM-MVC) method, which introduces a novel conceptualization of multiple views as a graph structure, and derives the minimum spanning tree of the views, reflecting the neighbouring relationships among them.
- 12
- Effective Video Abnormal Event Detection by Learning A Consistency-Aware High-Level Feature Extractor12
This work proposes a novel paradigm that performs VAD by learning a Consistency-Aware high-level Feature Extractor (CAFE), and designs a generic framework that can bridge high- level and low-level learning in VAD to further ameliorate VAD performance.
- Incomplete Multiview Clustering via Late Fusion12
A late fusion method for incomplete multiview clustering that performs kernel k-means clustering on the visible instances in each view and performs a late fusion of the clustering results from different views, which validates the effectiveness of the proposed method.
- 11
- Phrase Grounding-Based Style Transfer for Single-Domain Generalized Object Detection11
This paper proposes a novel phrase-grounding-based style transfer (PGST) approach for single-domain generalized object detection that significantly improves mean average precision (mAP) and outperforms or matches the performance of domain-adaptive object detection methods, which require target domain data for training, in several challenging scenarios.
- Automatically Identify and Rectify: Robust Deep Contrastive Multi-view Clustering in Noisy Scenarios11
A novel multi-view clustering framework for the automatic identification and rectification of noisy data, termed AIRMVC is proposed, which reformulate noisy identification as an anomaly identification problem using GMM and introduces a noise-robust contrastive mechanism to generate reliable representations.
- 11
- Incremental Nyström-based Multiple Kernel Clustering11
This work proposes a novel MKC algorithm, termed Incremental Nyström-based Multiple Kernel Clustering (INMKC), which employs a consensus clustering structure that aligns with the newly emerged base kernel matrix for updates, avoiding recalculating previous kernel matrices, thus saving substantial computational resources.
- 11
- 11
- 11
- 11
- 11
- Multi-View Deep Clustering based on AutoEncoder11
This work proposes a novel unsupervised deep embedding clustering multi-view method, which treats multiple different subspaces as different views through some data expansion methods for the same data.
- Absent extreme learning machine algorithm with application to packed executable identification11
By observing the fact that some structural characteristics of a part of packed malware instances hold unreasonable values, the proposed A-ELM algorithm cast the packed executable identification tasks into an absence learning problem, which can be efficiently addressed via the proposed A-ELM algorithm.
- 10
- Multi-view Graph Clustering with Dual Relation Optimization for Remote Sensing Data10
A novel Multi-view Graph Clustering with dual Relation Optimization (MDRO) framework tailored for remote sensing data is proposed, demonstrating that MDRO consistently outperforms existing single-view and multi-view clustering methods, achieving superior accuracy and robustness.
- 10
- 10
- Consensus Multiple Kernel K-Means Clustering With Late Fusion Alignment and Matrix-Induced Regularization10
The correlations between different clustering partitions are calculated as a matrix-induced regularization to encourage the diversity of clustering results and the proposed algorithm outperforms the existing state-of-the-art multi-kernel methods in clustering performance with less time complexity.
- 9
- 9
- 9
- Test-Time Training on Graphs with Large Language Models (LLMs)9
This paper designs a novel Test-Time Training pipeline, LLMTTT, which conducts the test-time adaptation under the annotations by LLMs on a carefully-selected node set and introduces a hybrid active node selection strategy that considers not only node diversity and representativeness, but also prediction signals from the pre-trained model.
- 9
- 9
- 9
- Adaptive Neighborhood Select Based on Local Linearity for Nonlinear Dimensionality Reduction9
Improved method of constructing neighborhood graph is presented, which can improve the accuracy of geodesic distance estimate for isometric embedding and improve the performance of most manifold algorithms, such as ISOMAP and LLE.
- 8
- 8
- 8
- 8
- Deep Incomplete Multi-View Clustering with Distribution Dual-Consistency Recovery Guidance8
This work proposes BURG, a novel method for incomplete multi-view clustering with distriBution dUalconsistency $\boldsymbol{R}$ ecovery $\boldsymbol{G}$ uidance, which treats each sample as a distinct category and performs cross-view distribution transfer to predict the distribution space of missing views.
- 8
- 8
- 8
- 8
- 8
- Multi-View Clustering With Learned Bipartite Graph8
Inspired by anchors used in spectral clustering on a bipartite graph, representative points that are updated during the iterating of MVC-LBG are introduced to improve the clustering accuracy and have lower computational complexity.
- 8
- 7
- 7
- 7
- Part-Aware Shape Generation With Latent 3D Diffusion of Neural Voxel Fields7
This article introduces a novel latent 3D diffusion model for generating neural voxel fields with precise part-aware structures and high-quality textures, and introduces a part-aware shape decoder to integrate the part codes into the neural voxel fields.
- 7
- 7
- Clustering then Propagation: Select Better Anchors for Knowledge Graph Embedding7
A novel anchor-based strategy for KGE, i.e., a relational clustering-based anchor selection strategy (RecPiece), where two characteristics are leveraged, i.e., representative ability of the cluster centroids and descriptive ability of relation types in KGs.
- A clustering ensemble algorithm for handling deep embeddings using cluster confidence7
A novel clustering ensemble algorithm for handling deep embeddings using cluster confidence (CEDECC) to improve the robustness and performance and consider the cluster-wise diversity and propose a novel ensemble cluster estimation to improve the overall consensus performance of the model.
- Rethinking SIGN Training: Provable Nonconvex Acceleration without First- and Second-Order Gradient Lipschitz7
This paper establishes the convergence of the sign-based method under weak first-order Lipschitz and proposes a relaxed second-order condition that still allows for nonconvex acceleration in sign- based methods.
- A Local Similarity-Preserving Framework for Nonlinear Dimensionality Reduction with Neural Networks7
Extensive experiments of data classification and clustering on eight real datasets show that Vec2vec is better than several classical dimensionality reduction methods in the statistical hypothesis test, and it is competitive with recently developed state-of-the-art UMAP.
- Simultaneous Clustering and Optimization for Evolving Datasets7
This paper proposes a new variant of the alternating direction method of multipliers (ADMM) to solve the problem of evolving datasets efficiently and extensive empirical studies confirm the effectiveness of this method.
- Accelerating SGD using flexible variance reduction on large-scale datasets7
A new flexible variance reduction method for SGD, named FVR-SGD, which can reduce memory overhead and speed up the convergence using flexible subset without extra computation is proposed, and linear convergence rate can be guaranteed while using flexible variance reduce.
- Spectral Discrepancy and Cross-Modal Semantic Consistency Learning for Object Detection in Hyperspectral Images6
A novel network termed Spectral Discrepancy and Cross-Modal semantic consistency learning (SDCM) is proposed, which facilitates the extraction of consistent information across a wide range of hyperspectral bands while utilizing the spectral dimension to pinpoint regions of interest.
- Find True Collaborators: Banzhaf Index-based Cross View Alignment for Partially View-aligned Clustering6
A Banzhaf Index-driven cross-view aligNment method, dubbed BIN, which systematically evaluates each view sample's contribution to joint decision-making within a game-theoretic framework and ensures robust matching in complex misalignment scenarios.
- 6
- 6
- Putting on the Thinking Hats: A Survey on Chain of Thought Fine-tuning from the Perspective of Human Reasoning Mechanism6
This survey presents the first comprehensive survey of CoT fine-tuning grounded in human reasoning theory, and is inspired by the well-known Six Thinking Hats framework, which systematically characterizes common human thinking modes using six metaphorical hats.
- Medical Federated Model with Mixture of Personalized and Sharing Components6
This paper proposes a new personalized framework of federated learning that successfully yields personalized models based on awareness of similarity between local data, and achieves better tradeoff between generalization and personalization than existing methods.
- 6
- 6
- Late Fusion Multi-view Clustering via Global and Local Alignment Maximization6
This paper reveals the theoretical connection of existing k-means clustering and the alignment between base partitions and the consensus one, and proposes a simple but effective multi-view algorithm termed LF-MVC-GAM, which optimally fuses multiple source information in partition level from each individual view and maximally aligns the consensus partition with these weighted base ones.
- 6
- Robustness Can Be Cheap: A Highly Efficient Approach to Discover Outliers under High Outlier Ratios6
A Low-rank based Efficient Outlier Detection (LEOD) framework to achieve favorable robustness against high outlier ratios with much cheaper computations to encourage satisfying robustness with low computational cost later is proposed.
- 6
- 5
- 5
- SparseMVC: Probing Cross-view Sparsity Variations for Multi-view Clustering5
Adaptive Sparse Autoencoders for M ulti-V iew C lustering (SparseMVC), a framework with three key modules that extends sparsity handling from the data-level to view-level and mitigating the adverse effects of encoding discrepancies through sample-level dynamic weighting.
- 5
- 5
- Generalized Deep Multi-View Clustering Via Causal Learning With Partially Aligned Cross-View Correspondence5
This paper is the first work to deal generalized multi-view clustering via causal learning via causal learning, and designs a Variational Auto-Encoder for causal learning by incorporating an encoder from existing information to estimate the invariant features.
- DGCBench: A Deep Graph Clustering Benchmark5
This work introduces DGCBench, the first comprehensive and unified benchmark for DGC methods, and develops PyDGC, an open-source Python library that standardizes the DGC training and evaluation paradigm.
- 5
- 5
- ProtoRefine: Enhancing Prototypes with Similar Structure in Few-Shot Learning5
This work introduces ProtoRefine, which employs the structural information of categories within the base domain that bear relevance to the new tasks, generating additional sample embeddings that refines the prototype representation, thus providing a more accurate prototype for category classification.
- 5
- 5
- 5
- Learning Vector-valued Functions with Local Rademacher Complexity and Unlabeled Data5
A unified learning framework for vector-valued functions based which is solved by proximal gradient descent on the primal is devised, achieving a much better tradeoff between accuracy and efficiency.
- 5
- Selecting the Optimal Hidden Layer of Extreme Learning Machine Using Multiple Kernel Learning5
The MKL method is utilized to adopt multiple kernel learning (MKL) to design a multi-hidden-layer-kernel ELM (MHLK-ELM), which achieves superior performance compared with conventional ELM and basic HLK- ELM.
- Multi-view Clustering with Noisy Views5
This work proposes a novel multi-view clustering method which assigns the noisy views with smaller weights to alleviate the negative effect of the noise, and develops an efficient algorithm, which updates partial variables alternatively.
- Model-aware categorical data embedding: a data-driven approach5
Extensive experiments of classification with the embedded representation on 17 data sets demonstrate that the proposed framework can significantly improve the categorical data representation performance compared with state-of-the-art competitors.
- Joint diversity regularization and graph regularization for multiple kernel k-means clustering via latent variables5
This paper develops and proposes a multiple kernel k-means clustering via latent variables (MKKLV) algorithm, in which base kernels can be adaptively adjusted with respect to each sample, and outperforms state-of-the-art techniques on several different benchmark datasets.
- Nonlinear discriminant clustering based on spectral regularization5
A novel nonlinear discriminant clustering by dimensional reduction based on spectral regularization is proposed that can obtain nonlinear low-dimensional representation that can recover the intrinsic manifold structure as well as enhance the cluster structure of the original high-dimensional data.
- “Stones From Other Hills Can Polish Jade”: Zero-Shot Anomaly Synthesis via Cross-Domain Anomaly Injection4
This work proposes a novel method named Cross-domain Anomaly Injection (CAI), which directly exploits cross-domain anomalies to enable highly authentic ZSAS in a training-free manner and builds the first Domain-agnostic Anomaly Dataset (DAAD) within the best knowledge, which provides ZSAS with abundant real anomaly patterns.
- 4
- AudioSet-R: A Refined AudioSet with Multi-Stage LLM Label Reannotation4
A three-stage reannotation framework that harnesses general-purpose audio-language foundation models to systematically improve the label quality of AudioSet and employs a cross-modal prompting strategy, inspired by the concept of prompt chaining.
- Learning the Anchors with Similar Distributions to Original Data for Multi-view Clustering4
This paper designs a Multi-View OT (MVOT) framework through complementary and consensus representation learning, and establishes the strong dual condition for MVOT by the relative interior, and successfully obtains the transport plan between anchors and original data for each view within linear computational complexity.
- 4
- LRGR: Self-Supervised Incomplete Multi-View Clustering via Local Refinement and Global Realignment4
The Local Refinement and Global Realignment (LRGR) Self-supervised model is proposed for incomplete multi-view clustering, which includes two stages, designed to enhance the versatility of imputed representations through cross-view contrastive learning guided by view-specific prototypes.
- Divide-Then-Rule: A Cluster-Driven Hierarchical Interpolator for Attribute-Missing Graphs4
Experimental results on 6 widely used graph datasets show that DTRGC significantly improves the clustering performance of various DGC methods under attribute-missing graphs.
- Spatially Resolved Transcriptomics Data Clustering with Tailored Spatial-scale Modulation4
This work introduces a novel spatially resolved transcriptomics data clustering network (TSstc), which surpasses existing state-of-the-art methods in clustering tasks and related downstream applications and uses a zero-inflated negative binomial model to capture the global probability distribution of gene expression profiles.
- Wigley Hull Design Optimization Based on Artificial Neural Network and Genetic Algorithm4
The results show that the resistance performance of the Wigley hull can be improved by designing the hull form reasonably and can be taken as a practical and efficient design tool.
- Scalable k-means for large-scale clustering4
This paper proposes two scalable k -means algorithms, Scalable Lloyd’s k-means and Scalable Mini-Batch k-means, distributed extensions of Lloyd’s algorithm and the mini-batch k -means, respectively that use the data-parallel technique to scale beyond computational and memory limits of a single machine.
- 3
- Graph Masked Autoencoder for Multi-view Remote Sensing Data Clustering3
Clustering-Guided graph Mask AutoEncoder (CG-MAE), the first framework to extend graph masked autoencoders to multi-view remote sensing clustering, is proposed, which introduces a clustering-guided masking strategy that selectively masks nodes near cluster centers and intra-cluster edges, which are crucial for capturing key structural information.
- Hierarchical Cross-View Alignment for Multi-View Clustering via Decoupled Information Distillation3
This work proposes HOARD, a novel framework for Hierarchical crOss-view Alignment for multi-view clusteRing via Decoupled information distillation that structurally decouples multi-view representations into shared and specific components, and performs hierarchical alignment.
- Game Theory Inspired Cross-View Interaction Alignment for Partially View-Aligned Clustering3
This work proposes a Banzhaf game theory-driven cross-view Interaction method, dubbed BIN, which fundamentally mitigates the reliance on pre-aligned samples and enables robust clustering even under severe view misalignment, and conducts extensive and in-depth experiments on multi-view data.
- 3
- 3
- Discriminative Binary Multi-View Clustering3
This work proposes a novel discriminative binary multi-view clustering (DBMVC) method that markedly outperforms other state-of-the-art methods in terms of effectiveness and efficiency.
- 3
- 3
- Multiple kernel clustering with late fusion consensus local graph preserving3
This paper proposes a late fusion MKC method with local graph refinement that unifies the traditional weighted multiple kernel k‐means, kernel partition, and graph construction into a single optimization procedure and enjoys more flexible kernel representations, leading to significant clustering improvements.
- 3
- Differentially Private ERM Based on Data Perturbation.3
This paper proposes a new `Data perturbation' based (DB) paradigm for DP-ERM: adding random noise to the original training data and achieving ($\epsilon,\delta$)-differential privacy on the final machine learning model, along with the preservation on the original data.
- Multiple Kernel Clustering Based on Self-Weighted Local Kernel Alignment3
A multiple kernel clustering algorithm based on self-weighted local kernel alignment, which can learn a proper weight to clustering performance for each local kernel alignment, is designed and tested.
- Robust Multi-View Clustering With a Unified Weight Learning Paradigm3
A novel robust multi-view clustering method with a unified weight learning paradigm that adaptively learns the quality of data points and the clustering capability of each view and improves the learning process of data quality.
- A Convergent Solution to Matrix Bidirectional Projection Based Feature Extraction with Application to Face Recognition3
A novel feature extraction method which exploits the Maximum Margin Criterion is proposed, where an iterative optimization algorithm is designed to compute the two projection matrices and it can monotonously increase the optimization objective.
- 2
- Embedded Multi-View Clustering via Fusing Tensor Subspace Representation and Adaptive Graph Learning2
- 2
- Balanced Multi-View Clustering2
A novel balanced multi-view clustering (BMvC) method is proposed, which introduces a view-specific contrastive regularization (VCR) to modulate the optimization of each view and achieves a better trade-off between the exploitation of view-specific patterns and the exploration of view-invariance patterns.
- 2
- SCAD: A self-constrained solution to automate context-guided zero-shot image anomaly detection2
This work proposes to automate context-guided ZS-IAD by a novel Self-Constrained Anomaly Detector (SCAD), which makes the following contributions: a novel self-constrained mechanism that can automatically determine proper values for key hyperparameters, and develops self-constrained normality refinement strategies that can actively constrain anomalies' impact and automatically rectify the stopping threshold.
- Sample-specific Modality Diagnosis and Cross-modal Enhancement for Incomplete Multimodal Representations2
Experiments show that SMCIR outperforms existing full-modal and non-recovery-based methods, well validating its efficacy and superiority in multimodal learning.
- Cross-level interaction and multi-granularity contrastive learning for multi-view clustering2
This work proposes a novel framework dubbed CLMGC: cross-level interaction and multi-granularity contrastive learning for multi-view clustering, which enhances information transfer between different levels and significantly improves the discriminative capability of the fine-grained semantic cluster, thus optimizing the overall performance of CLMGC.
- DPFMVC: Dynamic Progressive Fusion for Multi-view Clustering2
This work proposes a novel Dynamic Progressive Fusion Multi-View Clustering (DPFMVC), which employs a view-adaptive fusion mechanism that dynamically selects the most similar views, reducing conflicts between views while preserving complementary information.
- AdaPower: Specializing World Foundation Models for Predictive Manipulation2
This work introduces AdaPower, a lightweight adaptation framework that transforms general-purpose WFMs into specialist world models through two novel components: Temporal-Spatial Test-Time Training (TS-TTT) for inference-time adaptation and Memory Persistence (MP) for long-horizon consistency.
- SAINT: Sequence-Aware Integration for Spatial Transcriptomics Multi-View Clustering2
SAINT (Sequence-Aware Integration for Nucleotide-informed Transcriptomics), a framework that augments spatial representation learning with nucleotide-derived features, is proposed that consistently improves clustering performance across multiple datasets.
- 2
- SALVG: Latent Variable Gene Augmented Graph Learning for Multi-View Clustering in Spatial Transcriptomics2
SALVG (Spatial Augmentation via Latent Variable Genes), a novel and plug-and-play framework that leverages LVG-derived structural priors to enhance HVG representation learning for spatial clustering, is proposed.
- 2
- FS-KEN: Few-shot Knowledge Graph Reasoning by Adversarial Negative Enhancing2
This paper is the first to use GAN to conduct data augmentation on FS-KGR scenario, and proposes a novel algorithm based on adversarial learning for Enhancing Negative samples in few-shot scenarios of FS-KGR, termed FS-KEN.
- 2
- 2
- 2
- Variance Reduced K-Means Clustering2
A variance reduced k-mean VRKM is proposed, which outperforms the state-of-the-art method, and can be obtained 4× speedup for large-scale clustering.
- On Radius-Incorporated Multiple Kernel Learning2
Overall, radius-incorporated MKL algorithms achieve significant improvement over other counterparts in terms of classification performance, measured by classification accuracy and mean average precision.
- 1
- 1
- 1
- 1
- 1
- 1
- SpaCellAgent: A Self-Evolving LLM-Based Multi-Agent Framework for Trajectory Analysis1
This work proposes SpaCellAgent, an autonomous large language model (LLM) multi-agent framework that automates end-to-end spatiotemporal analysis and narrative generation and establishes a scalable, agent-driven paradigm for computational biology.
- VSumMamba: Mamba Empowered Efficient Video Summarization with Multi-Scale Spatial-Temporal Modeling1
This work introduces VSumMamba, an innovative video summarization approach that leverages Selective State Space Models to address the quadratic complexity limitations of Transformer based approaches meanwhile surpassing CNNs' restricted long-range modeling capabilities.
- Parameter-Free Clustering via Self-Supervised Consensus Maximization (Extended Version)1
This work proposes a novel and fully parameter-free clustering framework via Self-supervised Consensus Maximization, named SCMax, which performs hierarchical agglomerative clustering and cluster evaluation in a single, integrated process.
- Object Affordance Recognition and Grounding via Multi-scale Cross-modal Representation Learning1
A novel approach is proposed that learns an affordance-aware 3D representation and employs a stage-wise inference strategy leveraging the dependency between grounding and classification tasks, enabling inference of full potential affordance areas at a suitable regional scale.
- 1
- 1
- 1
- 1
- 1
- Revisiting Gradient Normalization and Clipping for Nonconvex SGD under Heavy-Tailed Noise: Necessity, Sufficiency, and Acceleration1
It is demonstrated that gradient normalization alone, assuming individual Lipschitz smoothness, is sufficient to ensure convergence of SGD under heavy-tailed noise, eliminating the need for gradient clipping.
- 1
- Meta-learning for Multi-variable Non-convex Optimization Problems: Iterating Non-optimums Makes Optimum Possible.1
A meta-learning based Global Scope Optimization (GSO) method that adaptively generates optimizers for sub-problems via meta-learners and constantly updates these meta- learners with respect to the global loss information of the overall problem.
- ℓ q -Norm Sample-Adaptive Multiple Kernel Learning1
A closed-form solution for optimizing the kernel weights based on the equivalence between group-lasso and MKL, and a framework for solving MKL problems in image classification is proposed.
- A framework for tuberculosis drug resistance classification based on deep transfer learning1
A transfer learning framework to classify the tuberculosis CT images which is provided in ImageCLEF 2017 challenge is adopted by firstly sampling 9 slices from 3D CT volumes and then training a deep transfer network based on VGG16, and finally voting to label the images.
- 1
- Sample-Based Extreme Learning Machine Regression with Absent Data1
This paper proposes a sample-based ELM regression algorithm, reformulated as a convex one, which can be readily implemented via off-the-shelf optimization packages and demonstrates superior performance over the compared ones, especially when the absent ratio is relatively intensive.
- 1
- Face Feature Extraction Based on Maximum Margin Criterion and Image Matrix Bidirectional Projection1
The proposed MMC-MBP approach can conquer the problem of existing tensor (matrix) projection based feature extraction approaches and outperforms the state-of-the-art methods in face feature extraction such as Eigenfaces, Fisherfaces, and Laplacianfaces.
- ANIMATE: Unsupervised Attributed Graph Anomaly Detection with Masked Graph Transformers–
Graph Transformers (GTs) are introduced into UGAD task, termed as unsupervised attributed graph Anomaly detectioN wIth Masked grAph TransformErs (ANIMATE), and can capture graph information that preserves the distinguishable characteristics of abnormalities from a global perspective.
- –
- –
- Learning Kernelized Hypothesis for Hidden Confounder Detection–
A Bi-level Kernel Confounder Detection (BiKCD) framework is contributed by learning adaptive kernelized structural space of structural functions by comparing coefficients from the higher-order regression and the classical ordinary least squares in learned kernelized space.
- –
- –
- –
- Beyond Parameter Finetuning: Test-Time Representation Refinement for Node Classification–
This work proposes TTReFT, a novel Test-Time Representation FineTuning framework that transitions the adaptation target from model parameters to latent representations to latent representations, and establishes representation finetuning as a new paradigm for graph TTT.
- Enhancing Kernel Power $K$-means: Scalable and Robust Clustering with Random Fourier Features and Possibilistic Method–
RFF-KPKM is proposed, introducing the first approximation theory for applying random Fourier features (RFF) to KPKM, an improved possibilistic RFF-based multiple kernel power k-means and improves robustness and ability to learn multiple kernels.
- –
- –
- Beyond Independent Manipulation: Individual Fairness-aware Strategic Classification with Peer Imitation–
Individual fairness-aware strategic classification (IFSC) is introduced, a framework that models peer-driven manipulation arising from individual fairness, where agents imitate nearby positively decided peers to obtain favorable outcomes.
- Self-Regressive Prototype Refinement: Stepping from Local to Global Prototypes in Few-Shot Image Classification–
A Self-Regressive Prototype Refinement (SRPR) is proposed to address the issue of incomplete prototype representations in few-shot learning and achieves state-of-the-art results across multiple few-shot benchmarks, particularly in the challenging 1-shot setting.
- A Survey on Human-Centric Voice–Face Multimodal Learning–
This survey provides a systematic overview of voice-face multimodal learning by categorizing research into five key areas, establishing the theoretical underpinnings of voice-face perception and providing insights for model design.
- CODE: Cross-Context Identity Extraction for Value Decomposition Based Multi-Agent Reinforcement Learning–
CODE is proposed, a generic identity-aware framework designed to enhance the cross-context generalization capability of value decomposition (VD) based MARL algorithms such as VDN, QMIX and QPLEX, strengthening the contextual adaptability of VD MARL algorithms.
- Self-Refining Spherical Consensus Embedding for Constrained Multi-View Clustering–
This work proposes a novel deep constrained multi-view clustering framework, namely SeSCE, which encodes view embeddings into a spherical space, leveraging pairwise constraints to maximize intra-class compactness and inter-class separability and introduces a confidence-aware pseudo-constraint mining mechanism.
- –
- –
- –
- Resisting Label Drift: Real-Time Multi-View Clustering with Semantic Consistency–
This work constructs a compact historical knowledge base via an adaptive diversity-aware selection mechanism, which guides the clustering of incoming data, enabling immediate inference without accessing the full history in a real-time semantic consistent incremental multi-view clustering framework.
- –
- –
- –
- AMGC2: Rethinking Deep Graph Clustering With a High Attribute-Missing Ratio–
A novel attribute-missing DGC framework called AMGC2 is proposed, which leverages the prediction of intra-cluster feature distribution to enhance robustness in clustering and data completion at a high attribute-missing ratio and performs cost-efficient prototype-level non-contrastive learning over completed samples.
- Prompt Learning for Textual Heterogeneous Information Networks–
This work is among the first to introduce and exploit the idea of prompt learning to align HIN features and textual features, and is composed of a text encoder and a graph encoder and uses contrastive learning to align and fuse the graph–text pair.
- –
- –
- Align Entities With Ontologies: LLM-Enhanced Inductive Subgraph Reasoning Over Ontology-Based Knowledge Graphs–
A subgraph-aware semantic expansion module that coordinates $k$-hop neighborhood information with LLM-generated descriptions to alleviate structural sparsity and a contrastive ontology matching mechanism resolves structural inconsistencies by computing adaptive similarity metrics between ontology embeddings and subgraph-derived semantic prototypes.
- Transport Barycenter-Guided Sample-to-Cluster Matching for Unaligned Multi-view Clustering–
Multi-view Discrete Optimal Transport is instantiated as a cluster-level transport matching framework that first adopts semi-discrete OT to learn a global OT barycenter via aggregating semantic-geometric information from all views, and then transports samples to barycenter-anchored consensus clusters under consistency constraints on semantic assignment and geometric measure.
- EAReranker: Efficient Embedding Adequacy Assessment for Retrieval Augmented Generation–
EAReranker is proposed, an efficient embedding-based adequacy assessment framework that evaluates document utility for RAG systems without requiring access to original text content and achieves competitive performance with state-of-the-art plaintext rerankers while maintaining constant memory usage.
- Scalable Cross-View Sample Alignment for Multi-View Clustering with View Structure Similarity–
This work proposes a scalable sample-alignment-based multi-view clustering method, referred to as SSA-MVC, which first employs a cluster-label matching (CLM) algorithm to select the view whose clustering labels best match those of the others as the benchmark view, and builds representations of non-aligned samples by computing their similarities with aligned samples.
- Bit-swapping Oriented Twin-memory Multi-view Clustering in Lifelong Incomplete Scenarios–
A named BSTM algorithm that synchronizes with the distinct dimensions by introducing a group of specialized projectors, and devise a shared bipartite graph construction via indicators to quantify similarity, which not only avoids redundant data-recalculations but alleviates the representation distortion caused by fusion.
- Intra-view and Inter-view Correlation Guided Multi-view Novel Class Discovery–
This paper proposes a novel framework named Intra-view and Interview Correlation Guided Multi-view Novel Class Discovery (IICMVNCD), which is the first attempt to explore NCD in multi-view setting so far and results validate the effectiveness of the proposed approach.
- –
- –
- –
- –
- Medical Federated Learning with Improved Representation and Personalized Aggregation–
This work proposes a method to extract centralized features meanwhile filtering random components, and conduct personalized aggregation, which achieves the best performance in metrics including accuracy, specificity, sensitivity, and F1 score.
- –
- –
- –
- –
- –
- –
- –
- –
- A Theoretical Revisit to Linear Convergence for Saddle Point Problems–
Convex-concave bilinear Saddle Point Problems (SPP) is widely used in lasso problems, Support Vector Machines, game theory, and so on.
- A Kernel Method with Manifold Regularization for Interactive Segmentation–
A kernel based semi-supervised learning framework with manifold regularization for interactive image segmentation with extensive experimental results demonstrate the effectiveness of the proposed approach.
- Optimization of Outsourcing ELM Problems in Cloud Computing from Multi-parties–
Theoretical analysis and experiments have shown that the proposed multi-parties cooperating mechanism for outsourcing ELM can effectively release customers from heavy computations.
- –
- A Max-Margin Learning Algorithm with Additional Features–
This paper shows how to use a max-margin learning framework to classify the incomplete data directly without any completing of the missing features, and develops an objective function which aims to maximize the margin of each sample in its own relevant subspace.
- A Large Margin Classifier with Additional Features–
It is shown how incomplete data can be classified directly without completing any missing features in a large-margin learning framework by avoiding the pre-processing phase in which the data is completed and the algorithm can achieve better or comparable classification accuracy compared to the existing algorithms.
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-12. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.