Accurate muscle-mass assessment is crucial for staging and managing sarcopenia, yet existing methods suffer from modality-specific limitations and weak integration of muscle function indicators. To solve these limitations, we propose a Dual-source Features Graph for Sarcopenia Evaluation (DFGSE) to synergize high- and low-energy whole-body Dual-energy X-ray Absorptiometry (DXA) images, local high-energy DXA images, and blood-borne biochemical markers. Specifically, the feature extraction module employs dual-energy feature extraction to disentangle soft-tissue and skeletal cues from low-energy images, while skeleton-aware detection extracts joint features from high-energy images. It yields global and local DXA embeddings, complemented by blood-test representations. In the relevance exploration module, inter- and intra-modality correlations are computed via bilinear transformations to form adjacency matrices for the global, local, and blood modality representations. These matrices seed the Multi-type Multi-relation Graph Convolutional Network (MMGCN) – the core of the relation learning module – which captures both direct and indirect interactions among modalities through relation-aware message passing. Finally, the graph-fused representations are used by a muscle-mass prediction head trained with cross-entropy loss. Experiments on the public MURA dataset and two independent sarcopenia cohorts demonstrate that DFGSE consistently outperforms machine learning and state-of-the-art graph-based methods, in terms of four evaluation metrics for classification task.

Unsupervised multi-view feature selection involves selecting a subset of crucial features across diverse views to diminish feature dimensionality without leveraging label information. While numerous studies have delved into this area, current solutions predominantly rely on linear multi-view data or employ weakly supervised learning to aid in feature selection. These approaches may risk losing semantic information when applied to real-world multi-view datasets. In this study, we introduce a novel model, Unsupervised Kernel-based Multi-view Feature selection with Robust self-representation and Binary hashing (UKMFS), which aims to identify robust consistent graph representation across views and leverage binary hashing codes to guide feature selection. Specifically, we first explore the underlying geometry by unifying the dimension of multi-view data with non-linear kernel mapping. Then, we search the consistent graph across views by fusing unique graph representations of each view in a self-representation manner. Additionally, we impose low-rank constraints on the graph of each view to mitigate noise and unimportant parts for preserving the main structures and patterns. Furthermore, we design an unsupervised hashing feature selection model to exploit reliable binary labels across views and weighted matrices from each view. Finally, an effective optimization method is customised to solve the formulated problem iteratively. Comprehensive experiments on public multi-view datasets indicate that our proposed method achieves state-of-the-art performance compared with the representative comparison methods regarding the clustering and the feature selection task.

Diffusion models have exhibited state-of-the-art performance in generative tasks across various domains. A few recent works leveraged the powerful modeling ability of the diffusion model to time-series forecasting, leading to a significant breakthrough. However, all these works perform the forecasting through incorporating the historical time-series conditions into the backward denoising. This causes the diffusion model to lose the essential consistency between forward and backward processes, thereby limiting the precision of the inference. In this paper, we propose a novel Dual Trajectory Revised Diffusion Model (TimeDTR) for time-series forecasting, which leverages an unconventional conditioning strategy to incorporate the historical information into both forward and backward trajectories in the diffusion model. Experimental results on six real-world datasets demonstrate that TimeDTR takes a big step forward from the state-of-the-art in time-series forecasting, especially in the long-term forecasting tasks, in terms of forecasting accuracy. The codes of the experiments with datasets and our algorithms are available at https://github.com/hhzzlll/TimeDTR.

Plasticity and stability denote the ability to assimilate new tasks while preserving previously acquired knowledge, representing two important concepts in continual learning. Recent research addresses stability by leveraging pre-trained models to provide informative representations, yet the efficacy of these methods is highly reliant on the choice of the pre-trained backbone, which may not yield optimal plasticity. This paper addresses this limitation by introducing a streamlined and potent framework that orchestrates multiple different pre-trained backbones to derive semantically rich multi-source representations. We propose an innovative Multi-Scale Interaction and Dynamic Fusion (MSIDF) technique to process and selectively capture the most relevant parts of multi-source features through a series of learnable attention modules, thereby helping to learn better decision boundaries to boost performance. Furthermore, we introduce a novel Multi-Level Representation Optimization (MLRO) strategy to adaptively refine the representation networks, offering adaptive representations that enhance plasticity. To mitigate over-regularization issues, we propose a novel Adaptive Regularization Optimization (ARO) method to manage and optimize a switch vector that selectively governs the updating process of each representation layer, which promotes the new task learning. The proposed MLRO and ARO approaches are collectively optimized within a unified optimization framework to achieve an optimal trade-off between plasticity and stability. Our extensive experimental evaluations reveal that the proposed framework attains state-of-the-art performance. The source code of our algorithm is available at https://github.com/CL-Coder236/LMSRR.

Continual learning requires the model to continually capture novel information without forgetting prior knowledge. Nonetheless, existing studies predominantly address the catastrophic forgetting, often neglecting enhancements in model robustness. Consequently, these methodologies fall short in real-time applications, such as autonomous driving, where data samples frequently exhibit noise due to environmental and lighting variations, thereby impairing model efficacy and causing safety issues. In this paper, we address robustness in continual learning systems by introducing an innovative approach, the Dynamic Siamese Expansion Framework (DSEF) that employs a Siamese backbone architecture, comprising static and dynamic components, to facilitate the learning of both global and local representations over time. Specifically, the proposed framework dynamically generates a lightweight expert for each novel task, leveraging the Siamese backbone to enable rapid adaptation. A novel Robust Dynamic Representation Optimization (RDRO) approach is proposed to incrementally update the dynamic backbone by maintaining all previously acquired representations and prediction patterns of historical experts, thereby fostering new task learning without inducing detrimental knowledge transfer. Additionally, we propose a novel Robust Feature Fusion (RFF) approach to incrementally amalgamate robust representations from all historical experts into the expert construction process. A novel mutual information-based technique is employed to derive adaptive weights for feature fusion by assessing the knowledge relevance between historical experts and the new task, thus maximizing positive knowledge transfer effects. A comprehensive experimental evaluation, benchmarking our approach against established baselines, demonstrates that our method achieves state-of-the-art performance even under adversarial attacks.

Extant studies predominantly address catastrophic forgetting within a simplified continual learning paradigm, typically confined to a singular data domain. Conversely, real-world applications frequently encompass multiple, evolving data domains, wherein models often struggle to retain many critical past information, thereby leading to performance degradation. This paper addresses this complex scenario by introducing a novel dynamic expansion approach called Learning Expandable and Adaptable Representations (LEAR). This framework orchestrates a collaborative backbone structure, comprising global and local backbones, designed to capture both general and task-specific representations. Leveraging this collaborative backbone, the proposed framework dynamically create a lightweight expert to delineate decision boundaries for each novel task, thereby facilitating the prediction process. To enhance new task learning, we introduce a novel Mutual Information-Based Prediction Alignment approach, which incrementally optimizes the global backbone via a mutual information metric, ensuring consistency in the prediction patterns of historical experts throughout the optimization phase. To mitigate network forgetting, we propose a Kullback–Leibler (KL) Divergence-Based Feature Alignment approach, which employs a probabilistic distance measure to prevent significant shifts in critical local representations. Furthermore, we introduce a novel Hilbert-Schmidt Independence Criterion (HSIC)-Based Collaborative Optimization approach, which encourages the local and global backbones to capture distinct semantic information in a collaborative manner, thereby mitigating information redundancy and enhancing model performance. Moreover, to accelerate new task learning, we propose a novel Expert Selection Mechanism that automatically identifies the most relevant expert based on data characteristics. This selected expert is then utilized to initialize a new expert, thereby fostering positive knowledge transfer. This approach also enables expert selection during the testing phase without requring any task information. Empirical results demonstrate that the proposed framework achieves state-of-the-art performance.

Multi-view feature selection across diverse views identifying a compact subset of the most informative feature across various data views without relying on labeled information. While most of the solutions are limited to linear multi-view data or utilize weakly-supervised single-label learning to assist in feature selection, leading to the loss of valuable semantic information, especially when dealing with complex real-world multi-view datasets. To overcome these limitations, we introduce a novel Resilient Kernel-based Unsupervised Multi-view Feature Selection via compact Binary Hashing (RKUMBH), which aims to search a robust and consistent graph representation across views, leveraging binary hashing codes to guide feature selection. Specifically, we first standardize the dimensionality of multi-view data by using non-linear kernel mapping. Then, we explore consistent graph structures across different views by fusing individual similarity graph of each view under a self-representation guidance. Moreover, the low-rank constraints are used to preserve the primary structures and patterns embedding within the data, and an unsupervised hashing feature selection framework is conducted to generate reliable hashing codes across views. Additionally, we design a customized iterative optimization method to solve the unified model. Extensive experiments on six public multi-view datasets demonstrate that our proposed method obtains state-of-the-art results compared to existing works for both clustering and feature selection tasks.

Due to network operation and maintenance relying heavily on network traffic monitoring, traffic matrix analysis has been one of the most crucial issues for network management related tasks. However, it is challenging to reliably obtain the precise measurement in computer networks because of the high measurement cost, and the unavoidable transmission loss. Although some methods proposed in recent years allowed estimating network traffic from partial flow-level or link-level measurements, they often perform poorly for traffic matrix estimation nowadays. Despite strong assumptions like low-rank structure and the prior distribution, existing techniques are usually task-specific and tend to be significantly worse as modern network communication is extremely complicated and dynamic. To address the dilemma, this paper proposed a diffusion-based traffic matrix analysis framework named Diffusion-TM, which leverages problem-agnostic diffusion to notably elevate the estimation performance in both traffic distribution and accuracy. The novel framework not only takes advantage of the powerful generative ability of diffusion models to produce realistic network traffic, but also leverages the denoising process to unbiasedly estimate all end-to-end traffic in a plug-and-play manner under theoretical guarantee. Moreover, taking into account that compiling an intact traffic dataset is usually infeasible, we also propose a two-stage training scheme to make our framework be insensitive to missing values in the dataset. With extensive experiments with real-world datasets, we illustrate the effectiveness of Diffusion-TM on several tasks. Moreover, the results also demonstrate that our method can obtain promising results even with 5% known values left in the datasets.

Unsupervised anomaly detection for multivariate time series (MTS) is a challenging task due to the difficulties of precisely learning the complex data patterns of MTS. The recent progress in sample generation achieved by diffusion models (DMs) motivates us to leverage the powerful learning ability of DMs to make a breakthrough in unsupervised anomaly detection for MTS. In this paper, we make the first attempt to design a novel diffusion-based anomaly detection model (named TimeADDM) for MTS data using the effective learning mechanism of DMs. To enhance the learning effect on MTS data, we propose to apply diffusion steps to the representations that accumulate the global time correlations through recurrent embedding. To enable the model for accurate anomaly detection, we design a reconstruction strategy that uses various levels of diffusion to compute the anomaly scores from different angles. By comparing TimeADDM with the state-of-the-art benchmarks, the results demonstrate that TimeADDM outperforms all baselines in terms of detection accuracy in four real-world MTS datasets and makes an improvement on the F1 score by up to 22%. The codes of the experiments with datasets and our algorithms are available at https://github.com/Hurongyao/TIMEADDM.

Current multiple object tracking methods typically associate two detected objects from consecutive frames via discriminative appearance features or motion modeling at the object level. However, in scenarios of dense crowds and prolonged occlusions, the extracted object-level features lack reliability, resulting in less effective target association. To tackle this challenge, we introduce GAT, a novel Group-Aware Transformer that learns to automatically group objects and complement targets’ appearance features with multi-level contextual information. In response to the prolonged occlusions issue, we further introduce an effective Trajectory Merging Mechanism (TMM), which relink the failed trajectories in prolonged occlusion based on motion patterns inferred from historical information. We demonstrate the effectiveness of our method with ablative experiments and exhibit outstanding tracking performance on the three popular multiple object tracking benchmarks MOT17, MOT20, and DanceTrack.

Abstract In computer-aided diagnostic systems, the functional connectome approach has become a common method for detecting neurological disorders. However, the existing methods either ignore the uniqueness of different subjects across the functional connectivities or neglect the commonality of the same disease for the functional connectivity of each subject, resulting in a lack of capacity of capturing a comprehensive functional model. To solve the issues, we develop a hierarchical graph learning with convolutional network that not only considers the unique information of each subject, but also takes the common information across subjects into account. Specifically, the proposed method consists of two structures, one is the individual graph model which selects the representative brain regions by combining each subject feature and its related brain region-based graph. The other is the population graph model to directly conduct classification performance by updating the information of each subject which considers both the subject itself and the nearest neighbours. Experimental results indicate that the proposed method on four real datasets outperforms the state-of-the-art approaches.

Graph convolutional network (GCN) outputs powerful representation by considering the structure information of the data to conduct representation learning, but its robustness is sensitive to the quality of both the feature matrix and the initial graph. In this article, we propose a novel multigraph fusion method to produce a high-quality graph and a low-dimensional space of original high-dimensional data for the GCN model. Specifically, the proposed method first extracts the common information and the complementary information among multiple local graphs to obtain a unified local graph, which is then fused with the global graph of the data to obtain the initial graph for the GCN model. As a result, the proposed method conducts the graph fusion process twice to simultaneously learn the low-dimensional space and the intrinsic graph structure of the data in a unified framework. Experimental results on real datasets demonstrated that our method outperformed the comparison methods in terms of classification tasks.

Graph neural networks (GNNs) conduct feature learning by taking into account the local structure preservation of the data to produce discriminative features, but need to address the following issues, i.e., 1) the initial graph containing faulty and missing edges often affect feature learning and 2) most GNN methods suffer from the issue of out-of-example since their training processes do not directly generate a prediction model to predict unseen data points. In this work, we propose a reverse GNN model to learn the graph from the intrinsic space of the original data points as well as to investigate a new out-of-sample extension method. As a result, the proposed method can output a high-quality graph to improve the quality of feature learning, while the new method of out-of-sample extension makes our reverse GNN method available for conducting supervised learning and semi-supervised learning. Experimental results on real-world datasets show that our method outputs competitive classification performance, compared to state-of-the-art methods, in terms of semi-supervised node classification, out-of-sample extension, random edge attack, link prediction, and image retrieval.

The traffic matrix estimation (TME) problem has been widely researched for decades of years. Recent progresses in deep generative models offer new opportunities to tackle TME problems in a more advanced way. In this paper, we leverage the powerful ability of denoising diffusion probabilistic models (DDPMs) on distribution learning, and for the first time adopt DDPM to address the TME problem. To ensure a good performance of DDPM on learning the distributions of TMs, we design a preprocessing module to reduce the dimensions of TMs while keeping the data variety of each OD flow. To improve the estimation accuracy, we parameterize the noise factors in DDPM and transform the TME problem into a gradient-descent optimization problem. Finally, we compared our method with the state-of-the-art TME methods using two real-world TM datasets, the experimental results strongly demonstrate the superiority of our method on both TM synthesis and TM estimation.

Abstract Sample correlations and feature relations are two pieces of information that are needed to be considered in the unsupervised feature selection, as labels are missing to guide model construction. Thus, we design a novel unsupervised feature selection scheme, in this paper, via considering the completed sample correlations and feature dependencies in a unified framework. Specifically, self-representation dependencies and graph construction are conducted to preserve and select the important neighbors for each sample in a comprehensive way. Besides, mutual information and sparse learning are designed to consider the correlations between features and to remove the informative features, respectively. Moreover, various constraints are constructed to automatically obtain the number of important neighbors and to conduct graph partition for the clustering task. Finally, we test the proposed method and verify the effectiveness and the robustness on eight data sets, comparing with nine state-of-the-art approaches with regard to three evaluation metrics for the clustering task.

Both data augmentation and contrastive loss are the key components of contrastive learning. In this paper, we design a new multi-view unsupervised graph representation learning method including adaptive data augmentation and multi-view contrastive learning, to address some issues of contrastive learning ignoring the information from feature space. Specifically, the adaptive data augmentation first builds a feature graph from the feature space, and then designs a deep graph learning model on the original representation and the topology graph to update the feature graph and the new representation. As a result, the adaptive data augmentation outputs multi-view information, which is fed into two GCNs to generate multi-view embedding features. Two kinds of contrastive losses are further designed on multi-view embedding features to explore the complementary information among the topology and feature graphs. Additionally, adaptive data augmentation and contrastive learning are embedded in a unified framework to form an end-to-end model. Experimental results verify the effectiveness of our proposed method, compared to state-of-the-art methods.

The functional connectomics study on resting state functional magnetic resonance imaging (rs-fMRI) data has become a popular way for early disease diagnosis. However, previous methods did not jointly consider the global patterns, the local patterns, and the temporal information of the blood-oxygen-level-dependent (BOLD) signals, thereby restricting the model effectiveness for early disease diagnosis. In this paper, we propose a new graph convolutional network (GCN) method to capture local and global patterns for conducting dynamically functional connectivity analysis. Specifically, we first employ the sliding window method to partition the original BOLD signals into multiple segments, aiming at achieving the dynamically functional connectivity analysis, and then design a multi-view node classification and a temporal graph classification to output two kinds of representations, which capture the temporally global patterns and the temporally local patterns, respectively. We further fuse these two kinds of representation by the weighted concatenation method whose effectiveness is experimentally proved as well. Experimental results on real datasets demonstrate the effectiveness of our method, compared to comparison methods on different classification tasks.

The key challenge of co-saliency detection is to extract discriminative features to distinguish the common salient foregrounds from backgrounds in a group of relevant images. In this paper, we propose a new co-saliency detection framework which includes two strategies to improve the discriminative ability of the features. Specifically, on one hand, we segment each image to semantic superpixel clusters as well as generate different scales/sizes of images for each input image by the VGG-16 model. Different scales capture different patterns of the images. As a result, multi-scale images can capture various patterns among all images by many kinds of perspectives. Second, we propose a new method of Graph Convolutional Network (GCN) to fine-tune the multi-scale features, aiming at capturing the common information among the features from all scales and the private or complementary information for the feature of each scale. Moreover, the proposed GCN method jointly conducts multi-scale feature fine-tune, graph learning, and feature learning in a unified framework. We evaluated our method on three benchmark data sets, compared to state-of-the-art co-saliency detection methods. Experimental results showed that our method outperformed all comparison methods in terms of different evaluation metrics.

Abstract Laplacian support vector machine (LapSVM) is an extremely popular classification method and relies on a small number of labels and a Laplacian regularization to complete the training of the support vector machine (SVM). However, the training of SVM model and Laplacian matrix construction are usually two independent process. Therefore, In this paper, we propose a new adaptive LapSVM method to realize semi-supervised learning with a primal solution. Specifically, the hinge loss of unlabelled data is considered to maximize the distance between unlabelled samples from different classes and the process of dealing with labelled data are similar to other LapSVM methods. Besides, the proposed method embeds the Laplacian matrix acquisition into the SVM training process to improve the effectiveness of Laplacian matrix and the accuracy of new SVM model. Moreover, a novel optimization algorithm considering primal solver is proposed to our adaptive LapSVM model. Experimental results showed that our method outperformed all comparison methods in terms of different evaluation metrics on both real datasets and synthetic datasets.

The functional connectomic profile is one of the non-invasive imaging biomarkers in the computer-assisted diagnostic system for many neuro-diseases. However, the diagnostic power of functional connectivity is challenged by mixed frequency-specific neuronal oscillations in the brain, which makes the single Functional Connectivity Network (FCN) often underpowered to capture the disease-related functional patterns. To address this challenge, we propose a novel functional connectivity analysis framework to conduct joint feature learning and personalized disease diagnosis, in a semi-supervised manner, aiming at focusing on putative multi-band functional connectivity biomarkers from functional neuroimaging data. Specifically, we first decompose the Blood Oxygenation Level Dependent (BOLD) signals into multiple frequency bands by the discrete wavelet transform, and then cast the alignment of all fully-connected FCNs derived from multiple frequency bands into a parameter-free multi-band fusion model. The proposed fusion model fuses all fully-connected FCNs to obtain a sparsely-connected FCN (sparse FCN for short) for each individual subject, as well as lets each sparse FCN be close to its neighbored sparse FCNs and be far away from its furthest sparse FCNs. Furthermore, we employ the$\ell _{{1}}$-SVM to conduct joint brain region selection and disease diagnosis. Finally, we evaluate the effectiveness of our proposed framework on various neuro-diseases,i.e.,Fronto-Temporal Dementia (FTD), Obsessive-Compulsive Disorder (OCD), and Alzheimer’s Disease (AD), and the experimental results demonstrate that our framework shows more reasonable results, compared to state-of-the-art methods, in terms of classification performance and the selected brain regions. The source code can be visited by the urlhttps://github.com/reynard-hu/mbbna.

Brain functional connectivity analysis on fMRI data could improve the understanding of human brain function. However, due to the influence of the inter-subject variability and the heterogeneity across subjects, previous methods of functional connectivity analysis are often insufficient in capturing disease-related representation so that decreasing disease diagnosis performance. In this paper, we first propose a new multi-graph fusion framework to fine-tune the original representation derived from Pearson correlation analysis, and then employ L1-SVM on fine-tuned representations to conduct joint brain region selection and disease diagnosis for avoiding the issue of the curse of dimensionality on high-dimensional data. The multi-graph fusion framework automatically learns the connectivity number for every node (i.e., brain region) and integrates all subjects in a unified framework to output homogenous and discriminative representations of all subjects. Experimental results on two real data sets, i.e., fronto-temporal dementia (FTD) and obsessive-compulsive disorder (OCD), verified the effectiveness of our proposed framework, compared to state-of-the-art methods.

Previous multi-view spectral clustering methods are a two-step strategy, which first learns a fixed common representation (or common affinity matrix) of all the views from original data and then conducts k-means clustering on the resulting common affinity matrix. The two-step strategy is not able to output reasonable clustering performance since the goal of the first step (i.e., the common affinity matrix learning) is not designed for achieving the optimal clustering result. Moreover, the two-step strategy learns the common affinity matrix from original data, which often contain noise and redundancy to influence the quality of the common affinity matrix. To address these issues, in this paper, we design a novel One-step Multi-view Spectral Clustering (OMSC) method to output the common affinity matrix as the final clustering result. In the proposed method, the goal of the common affinity matrix learning is designed to achieving optimal clustering result and the common affinity matrix is learned from low-dimensional data where the noise and redundancy of original high-dimensional data have been removed. We further propose an iterative optimization method to fast solve the proposed objective function. Experimental results on both synthetic datasets and public datasets validated the effectiveness of our proposed method, comparing to the state-of-the-art methods for multi-view clustering.

This paper proposes a new unsupervised spectral feature selection method to preserve both the local and global structure of the features as well as the samples. Specifically, our method uses the self-expressiveness of the features to represent each feature by other features for preserving the local structure of features, and a low-rank constraint on the weight matrix to preserve the global structure among samples as well as features. Our method also proposes to learn the graph matrix measuring the similarity of samples for preserving the local structure among samples. Furthermore, we propose a new optimization algorithm to the resulting objective function, which iteratively updates the graph matrix and the intrinsic space so that collaboratively improving each of them. Experimental analysis on 12 benchmark datasets showed that the proposed method outperformed the state-of-the-art feature selection methods in terms of classification performance.

This paper proposes a one-step spectral clustering method by learning an intrinsic affinity matrix (i.e., the clustering result) from the low-dimensional space (i.e., intrinsic subspace) of original data. Specifically, the intrinsic affinitymatrix is learnt by: 1) the alignment of the initial affinity matrix learnt from original data; 2) the adjustment of the transformation matrix, which transfers the original feature space into its intrinsic subspace by simultaneously conducting feature selection and subspace learning; and 3) the clustering result constraint, i.e., the graph constructed by the intrinsic affinity matrix has exact c connected components where c is the number of clusters. In this way, two affinity matrices and a transformation matrix are iteratively updated until achieving their individual optimum, so that these two affinity matrices are consistent and the intrinsic subspace is learnt via the transformation matrix. Experimental results on both synthetic and benchmark datasets verified that our proposed method outputted more effective clustering result than the previous clustering methods.

Current unsupervised feature selection (UFS) methods learn the similarity matrix by using a simple graph which is learnt from the original data as well as is independent from the process of feature selection, and thus unable to efficiently remove the redundant/irrelevant features. To address these issues, we propose a new UFS method to jointly learn the similarity matrix and conduct both subspace learning (via learning a dynamic hypergraph) and feature selection (via a sparsity constraint). As a result, we reduce the feature dimensions using different methods (i.e., subspace learning and feature selection) from different feature spaces, and thus makes our method select the informative features effectively and robustly. We tested our method using benchmark datasets to conduct the clustering tasks using the selected features, and the experimental results show that our proposed method outperforms all the comparison methods.