Pseudo-Supervised Deep Subspace Clustering

Multimodal active subspace analysis for computing assessment oriented subspaces from neuroimaging data

BACKGROUND

For successful biomarker discovery, it is essential to develop computational frameworks that summarize high-dimensional neuroimaging data in terms of involved sub-systems of the brain, while also revealing underlying heterogeneous functional and structural changes covarying with specific cognitive and biological traits. However, unsupervised decompositions do not inculcate clinical assessment information, while supervised approaches extract only individual feature importance, thereby impeding qualitative interpretation at the level of subspaces.

NEW METHOD

We present a novel framework to extract robust multimodal brain subspaces associated with changes in a given cognitive or biological trait. Our approach involves active subspace learning on the gradients of a trained machine learning model followed by clustering to extract and summarize the most salient and consistent subspaces associated with the target variable.

RESULTS

Through a rigorous cross-validation procedure on an Alzheimer's disease (AD) dataset, our framework successfully extracts multimodal subspaces specific to a given clinical assessment (e.g., memory and other cognitive skills), and also retains predictive performance in standard machine learning algorithms. We also show that the salient active subspace directions occur consistently across randomly sub-sampled repetitions of the analysis.

COMPARISON WITH EXISTING METHOD(S)

Compared to existing unsupervised decompositions based on principle component analysis, the subspace components in our framework retain higher predictive information.

CONCLUSIONS

As an important step towards biomarker discovery, our framework not only uncovers AD-related brain regions in the associated brain subspaces, but also enables automated identification of multiple underlying structural and functional sub-systems of the brain that collectively characterize changes in memory and proficiency in cognitive skills related to brain disorders like AD.

Deep Fusion Clustering Network With Reliable Structure Preservation

Deep clustering, which can elegantly exploit data representation to seek a partition of the samples, has attracted intensive attention. Recently, combining auto-encoder (AE) with graph neural networks (GNNs) has accomplished excellent performance by introducing structural information implied among data in clustering tasks. However, we observe that there are some limitations of most existing works: 1) in practical graph datasets, there exist some noisy or inaccurate connections among nodes, which would confuse network learning and cause biased representations, thus leading to unsatisfied clustering performance; 2) lacking dynamic information fusion module to carefully combine and refine the node attributes and the graph structural information to learn more consistent representations; and 3) failing to exploit the two separated views' information for generating a more robust target distribution. To solve these problems, we propose a novel method termed deep fusion clustering network with reliable structure preservation (DFCN-RSP). Specifically, the random walk mechanism is introduced to boost the reliability of the original graph structure by measuring localized structure similarities among nodes. It can simultaneously filter out noisy connections and supplement reliable connections in the original graph. Moreover, we provide a transformer-based graph auto-encoder (TGAE) that can use a self-attention mechanism with the localized structure similarity information to fine-tune the fused topology structure among nodes layer by layer. Furthermore, we provide a dynamic cross-modality fusion strategy to combine the representations learned from both TGAE and AE. Also, we design a triplet self-supervision strategy and a target distribution generation measure to explore the cross-modality information. The experimental results on five public benchmark datasets reflect that DFCN-RSP is more competitive than the state-of-the-art deep clustering algorithms. The corresponding code is available at https://github.com/gongleii/DFCN-RSP.

Partial Tubal Nuclear Norm-Regularized Multiview Subspace Learning

In this article, a unified multiview subspace learning model, called partial tubal nuclear norm-regularized multiview subspace learning (PTN2MSL), was proposed for unsupervised multiview subspace clustering (MVSC), semisupervised MVSC, and multiview dimension reduction. Unlike most of the existing methods which treat the above three related tasks independently, PTN2MSL integrates the projection learning and the low-rank tensor representation to promote each other and mine their underlying correlations. Moreover, instead of minimizing the tensor nuclear norm which treats all singular values equally and neglects their differences, PTN2MSL develops the partial tubal nuclear norm (PTNN) as a better alternative solution by minimizing the partial sum of tubal singular values. The PTN2MSL method was applied to the above three multiview subspace learning tasks. It demonstrated that these tasks organically benefited from each other and PTN2MSL has achieved better performance in comparison to state-of-the-art methods.

Sampling clustering based on multi-view attribute structural relations

In light of the exponential growth in information volume, the significance of graph data has intensified. Graph clustering plays a pivotal role in graph data processing by jointly modeling the graph structure and node attributes. Notably, the practical significance of multi-view graph clustering is heightened due to the presence of diverse relationships within real-world graph data. Nonetheless, prevailing graph clustering techniques, predominantly grounded in deep learning neural networks, face challenges in effectively handling multi-view graph data. These challenges include the incapability to concurrently explore the relationships between multiple view structures and node attributes, as well as difficulties in processing multi-view graph data with varying features. To tackle these issues, this research proposes a straightforward yet effective multi-view graph clustering approach known as SLMGC. This approach uses graph filtering to filter noise, reduces computational complexity by extracting samples based on node importance, enhances clustering representations through graph contrastive regularization, and achieves the final clustering outcomes using a self-training clustering algorithm. Notably, unlike neural network algorithms, this approach avoids the need for intricate parameter settings. Comprehensive experiments validate the supremacy of the SLMGC approach in multi-view graph clustering endeavors when contrasted with prevailing deep neural network techniques.

Riemannian Manifold-Based Feature Space and Corresponding Image Clustering Algorithms

Image feature representation is a key factor influencing the accuracy of clustering. Traditional point-based feature spaces represent spectral features of an image independently and introduce spatial relationships of pixels in the image domain to enhance the contextual information expression ability. Mapping-based feature spaces aim to preserve the structure information, but the complex computation and the unexplainability of image features have a great impact on their applications. To this end, we propose an explicit feature space called Riemannian manifold feature space (RMFS) to present the contextual information in a unified way. First, the Gaussian probability distribution function (pdf) is introduced to characterize the features of a pixel in its neighborhood system in the image domain. Then, the feature-related pdfs are mapped to a Riemannian manifold, which constructs the proposed RMFS. In RMFS, a point can express the complex contextual information of corresponding pixel in the image domain, and pixels representing the same object are linearly distributed. This gives us a chance to convert nonlinear image segmentation problems to linear computation. To verify the superiority of the expression ability of the proposed RMFS, a linear clustering algorithm and a fuzzy linear clustering algorithm are proposed. Experimental results show that the proposed RMFS-based algorithms outperform their counterparts in the spectral feature space and the RMFS-based ones without the linear distribution characteristics. This indicates that the RMFS can better express features of an image than spectral feature space, and the expressed features can be easily used to construct linear segmentation models.

Collaborative Decision-Reinforced Self-Supervision for Attributed Graph Clustering

Attributed graph clustering aims to partition nodes of a graph structure into different groups. Recent works usually use variational graph autoencoder (VGAE) to make the node representations obey a specific distribution. Although they have shown promising results, how to introduce supervised information to guide the representation learning of graph nodes and improve clustering performance is still an open problem. In this article, we propose a Collaborative Decision-Reinforced Self-Supervision (CDRS) method to solve the problem, in which a pseudo node classification task collaborates with the clustering task to enhance the representation learning of graph nodes. First, a transformation module is used to enable end-to-end training of existing methods based on VGAE. Second, the pseudo node classification task is introduced into the network through multitask learning to make classification decisions for graph nodes. The graph nodes that have consistent decisions on clustering and pseudo node classification are added to a pseudo-label set, which can provide fruitful self-supervision for subsequent training. This pseudo-label set is gradually augmented during training, thus reinforcing the generalization capability of the network. Finally, we investigate different sorting strategies to further improve the quality of the pseudo-label set. Extensive experiments on multiple datasets show that the proposed method achieves outstanding performance compared with state-of-the-art methods. Our code is available at https://github.com/Jillian555/TNNLS_CDRS.

Contrastive self-representation learning for data clustering

This paper is concerned with self-representation subspace learning. It is one of the most representative subspace techniques, which has attracted considerable attention for clustering due to its good performance. Among these methods, low-rank representation (LRR) has achieved impressive results for subspace clustering. However, it only considers the similarity between the data itself, while neglecting the differences with other samples. Besides, it cannot well deal with noise and portray cluster-to-cluster relationships well. To solve these problems, we propose a Contrastive Self-representation model for Clustering (CSC). CSC simultaneously takes into account the similarity/dissimilarity between positive/negative pairs when learning the self-representation coefficient matrix of data while the form of the loss function can reduce the effect of noise on the results. Moreover, We use the ℓ1,2-norm regularizer on the coefficient matrix to achieve its sparsity to better characterize the cluster structure. Thus, the learned self-representation coefficient matrix well encodes both the discriminative information and cluster structure. Extensive experiments on seven benchmark databases indicate the superiority of our proposed method.

Self-Supervised Information Bottleneck for Deep Multi-View Subspace Clustering

In this paper, we explore the problem of deep multi-view subspace clustering framework from an information-theoretic point of view. We extend the traditional information bottleneck principle to learn common information among different views in a self-supervised manner, and accordingly establish a new framework called Self-supervised Information Bottleneck based Multi-view Subspace Clustering (SIB-MSC). Inheriting the advantages from information bottleneck, SIB-MSC can learn a latent space for each view to capture common information among the latent representations of different views by removing superfluous information from the view itself while retaining sufficient information for the latent representations of other views. Actually, the latent representation of each view provides a kind of self-supervised signal for training the latent representations of other views. Moreover, SIB-MSC attempts to disengage the other latent space for each view to capture the view-specific information by introducing mutual information based regularization terms, so as to further improve the performance of multi-view subspace clustering. Extensive experiments on real-world multi-view data demonstrate that our method achieves superior performance over the related state-of-the-art methods.

Graph Embedding Contrastive Multi-Modal Representation Learning for Clustering

Multi-modal clustering (MMC) aims to explore complementary information from diverse modalities for clustering performance facilitating. This article studies challenging problems in MMC methods based on deep neural networks. On one hand, most existing methods lack a unified objective to simultaneously learn the inter- and intra-modality consistency, resulting in a limited representation learning capacity. On the other hand, most existing processes are modeled for a finite sample set and cannot handle out-of-sample data. To handle the above two challenges, we propose a novel Graph Embedding Contrastive Multi-modal Clustering network (GECMC), which treats the representation learning and multi-modal clustering as two sides of one coin rather than two separate problems. In brief, we specifically design a contrastive loss by benefiting from pseudo-labels to explore consistency across modalities. Thus, GECMC shows an effective way to maximize the similarities of intra-cluster representations while minimizing the similarities of inter-cluster representations at both inter- and intra-modality levels. So, the clustering and representation learning interact and jointly evolve in a co-training framework. After that, we build a clustering layer parameterized with cluster centroids, showing that GECMC can learn the clustering labels with given samples and handle out-of-sample data. GECMC yields superior results than 14 competitive methods on four challenging datasets. Codes and datasets are available: https://github.com/xdweixia/GECMC.

SPICE: Semantic Pseudo-Labeling for Image Clustering

The similarity among samples and the discrepancy among clusters are two crucial aspects of image clustering. However, current deep clustering methods suffer from inaccurate estimation of either feature similarity or semantic discrepancy. In this paper, we present a Semantic Pseudo-labeling-based Image ClustEring (SPICE) framework, which divides the clustering network into a feature model for measuring the instance-level similarity and a clustering head for identifying the cluster-level discrepancy. We design two semantics-aware pseudo-labeling algorithms, prototype pseudo-labeling and reliable pseudo-labeling, which enable accurate and reliable self-supervision over clustering. Without using any ground-truth label, we optimize the clustering network in three stages: 1) train the feature model through contrastive learning to measure the instance similarity; 2) train the clustering head with the prototype pseudo-labeling algorithm to identify cluster semantics; and 3) jointly train the feature model and clustering head with the reliable pseudo-labeling algorithm to improve the clustering performance. Extensive experimental results demonstrate that SPICE achieves significant improvements (~10%) over existing methods and establishes the new state-of-the-art clustering results on six balanced benchmark datasets in terms of three popular metrics. Importantly, SPICE significantly reduces the gap between unsupervised and fully-supervised classification; e.g. there is only 2% (91.8% vs 93.8%) accuracy difference on CIFAR-10. Our code is made publicly available at https://github.com/niuchuangnn/SPICE.

Image Clustering Algorithm Based on Predefined Evenly-Distributed Class Centroids and Composite Cosine Distance

The clustering algorithms based on deep neural network perform clustering by obtaining the optimal feature representation. However, in the face of complex natural images, the cluster accuracy of existing clustering algorithms is still relatively low. This paper presents an image clustering algorithm based on predefined evenly-distributed class centroids (PEDCC) and composite cosine distance. Compared with the current popular auto-encoder structure, we design an encoder-only network structure with normalized latent features, and two effective loss functions in latent feature space by replacing the Euclidean distance with a composite cosine distance. We find that (1) contrastive learning plays a key role in the clustering algorithm and greatly improves the quality of learning latent features; (2) compared with the Euclidean distance, the composite cosine distance can be more suitable for the normalized latent features and PEDCC-based Maximum Mean Discrepancy (MMD) loss function; and (3) for complex natural images, a self-supervised pretrained model can be used to effectively improve clustering performance. Several experiments have been carried out on six common data sets, MNIST, Fashion-MNIST, COIL20, CIFAR-10, STL-10 and ImageNet-10. Experimental results show that our method achieves the best clustering effect compared with other latest clustering algorithms.

SVR-based Multimodal Active Subspace Analysis for the Brain using Neuroimaging Data.. [DOI: 10.1101/2022.07.28.501879] [Citation(s) in RCA: 0] [Impact Index Per Article: 0]

Understanding the patterns of changes in brain function and structure due to various disorders and diseases is of utmost importance. There have been numerous efforts toward successful biomarker discovery for complex brain disorders by evaluating neuroimaging datasets with novel analytical frameworks. However, due to the multi-faceted nature of the disorders involving a wide and overlapping range of symptoms as well as complex changes in structural and functional brain networks, it is increasingly important to devise computational frameworks that can consider the underlying patterns of heterogeneous changes with specific target assessments, at the same time producing a summarizing output from the high-dimensional neuroimaging data. While various machine learning approaches focus on diagnostic prediction, many learning frameworks analyze important features at the level of brain regions involved in prediction using supervised methods. Unsupervised learning methods have also been utilized to break down the neuroimaging features into lower dimensional components. However, most learning frameworks either do not consider the target assessment information while extracting brain subspaces, or can extract only higher dimensional importance associations as an ordered list of involved features, making manual interpretation at the level of subspaces difficult. We present a novel multimodal active subspace learning framework to understand various subspaces within the brain that are associated with changes in particular biological and cognitive traits. For a given cognitive or biological trait, our framework performs a decomposition of the feature importances to extract robust multimodal subspaces that define the most significant change in the given trait. Through a rigorous cross-validation procedure on an Alzheimer’s disease (AD) dataset, we show that our framework can extract subspaces covering both functional and structural modalities, which are specific to a given clinical assessment (like memory and other cognitive skills) and also retain predictive performance in standard machine learning algorithms. We show that our framework not only uncovers AD-related brain regions (e.g., hippocampus, entorhinal cortex) in the associated brain subspaces, but also enables an automated identification of multiple underlying structural and functional sub-systems of the brain that collectively characterize changes in memory and cognitive skill proficiency related to brain disorders like AD.

Multiview Spectral Clustering With Bipartite Graph

Multi-view spectral clustering has become appealing due to its good performance in capturing the correlations among all views. However, on one hand, many existing methods usually require a quadratic or cubic complexity for graph construction or eigenvalue decomposition of Laplacian matrix; on the other hand, they are inefficient and unbearable burden to be applied to large scale data sets, which can be easily obtained in the era of big data. Moreover, the existing methods cannot encode the complementary information between adjacency matrices, i.e., similarity graphs of views and the low-rank spatial structure of adjacency matrix of each view. To address these limitations, we develop a novel multi-view spectral clustering model. Our model well encodes the complementary information by Schatten p -norm regularization on the third tensor whose lateral slices are composed of the adjacency matrices of the corresponding views. To further improve the computational efficiency, we leverage anchor graphs of views instead of full adjacency matrices of the corresponding views, and then present a fast model that encodes the complementary information embedded in anchor graphs of views by Schatten p -norm regularization on the tensor bipartite graph. Finally, an efficient alternating algorithm is derived to optimize our model. The constructed sequence was proved to converge to the stationary KKT point. Extensive experimental results indicate that our method has good performance.

Deep structure integrative representation of multi-omics data for cancer subtyping

MOTIVATION

Cancer is a heterogeneous group of diseases. Cancer subtyping is crucial and critical step to diagnosis, prognosis and treatment. Since high-throughput sequencing technologies provide unprecedented opportunity to rapid collect multi-omics data for the same individuals, an urgent need in current is how to effectively represent and integrate these multi-omics data to achieve clinically meaningful cancer subtyping.

RESULTS

We propose a novel deep learning model, called Deep Structure Integrative Representation (DSIR), for cancer subtypes dentification by integrating representation and clustering multi-omics data. DSIR simultaneously captures the global structures in sparse subspace and local structures in manifold subspace from multi-omics data and constructs consensus similarity matrix by utilizing deep neural networks. Extensive tests are performed in twelve different cancers on three levels of omics data from The Cancer Genome Atlas. The results demonstrate that DSIR obtains more significant performances than the state-of-the-art integrative methods.

AVAILABILITY

https://github.com/Polytech-bioinf/Deep-structure-integrative-representation.git.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Adaptive Attribute and Structure Subspace Clustering Network

Deep self-expressiveness-based subspace clustering methods have demonstrated effectiveness. However, existing works only consider the attribute information to conduct the self-expressiveness, limiting the clustering performance. In this paper, we propose a novel adaptive attribute and structure subspace clustering network (AASSC-Net) to simultaneously consider the attribute and structure information in an adaptive graph fusion manner. Specifically, we first exploit an auto-encoder to represent input data samples with latent features for the construction of an attribute matrix. We also construct a mixed signed and symmetric structure matrix to capture the local geometric structure underlying data samples. Then, we perform self-expressiveness on the constructed attribute and structure matrices to learn their affinity graphs separately. Finally, we design a novel attention-based fusion module to adaptively leverage these two affinity graphs to construct a more discriminative affinity graph. Extensive experimental results on commonly used benchmark datasets demonstrate that our AASSC-Net significantly outperforms state-of-the-art methods. In addition, we conduct comprehensive ablation studies to discuss the effectiveness of the designed modules. The code is publicly available at https://github.com/ZhihaoPENG-CityU/AASSC-Net.

Efficient correntropy-based multi-view clustering with anchor graph embedding

Although multi-view clustering has received widespread attention due to its far superior performance to single-view clustering, it still faces the following issues: (1) high computational cost, considering the introduction of multi-view information, reduces the clustering efficiency greatly; (2) complex noises and outliers, existed in real-world data, pose a huge challenge to the robustness of clustering algorithms. Currently, how to increase the efficiency and robustness has become two important issues of multi-view clustering. To cope with the above issues, an efficient correntropy-based multi-view clustering algorithm (ECMC) is proposed in this paper, which can not only improve clustering efficiency by constructing embedded anchor graph and utilizing nonnegative matrix factorization (NMF), but also enhance the robustness by exploring correntropy to suppress various noises and outliers. To further improve clustering efficiency, one of the factors of NMF is constrained to be an indicator matrix instead of a traditional non-negative matrix, so that the categories of samples can be obtained directly without any extra operation. Subsequently, a novel half-quadratic-based strategy is proposed to optimize the non-convex objective function of ECMC. Finally, extensive experiments on eight real-world datasets and eighteen noisy datasets show that ECMC can guarantee faster speed and better robustness than other state-of-the-art multi-view clustering algorithms.

Multi-view graph embedding clustering network: Joint self-supervision and block diagonal representation

Multi-view clustering has become an active topic in artificial intelligence. Yet, similar investigation for graph-structured data clustering has been absent so far. To fill this gap, we present a Multi-View Graph embedding Clustering network (MVGC). Specifically, unlike traditional multi-view construction methods, which are only suitable to describe Euclidean structure data, we leverage Euler transform to augment the node attribute, as a new view descriptor, for non-Euclidean structure data. Meanwhile, we impose block diagonal representation constraint, which is measured by the ℓ_1,2-norm, on self-expression coefficient matrix to well explore the cluster structure. By doing so, the learned view-consensus coefficient matrix well encodes the discriminative information. Moreover, we make use of the learned clustering labels to guide the learnings of node representation and coefficient matrix, where the latter is used in turn to conduct the subsequent clustering. In this way, clustering and representation learning are seamlessly connected, with the aim to achieve better clustering performance. Extensive experimental results indicate that MVGC is superior to 11 state-of-the-art methods on four benchmark datasets. In particular, MVGC achieves an Accuracy of 96.17% (53.31%) on the ACM (IMDB) dataset, which is an up to 2.85% (1.97%) clustering performance improvement compared with the strongest baseline.