Low-rank Embedding Research Articles

Drug Target Prediction by Multi-View Low Rank Embedding.

Drug repositioning has been a key problem in drug development, and heterogeneous data sources are used to predict drug-target interactions by different approaches. However, most of studies focus on a single representation of drugs or proteins. It has been shown that integrating multi-view representations of drugs and proteins can strengthen the prediction ability. For example, a drug can be represented by its chemical structure, or by its chemical response in different cells. A protein can be represented by its sequence, or by its gene expression values in different cells. The docking of drugs and proteins based on their structure can be considered as one view (structural view), and the chemical performance of them based on gene expression and drug response can be considered as another view (chemical view). In this work, we first propose a single-view approach of SLRE based on low rank embedding for an arbitrary view, and then extend it to a multi-view approach of MLRE, which could integrate both views. Our experiments show that our methods perform significantly better than baseline methods including single-view methods and multi-view methods. We finally report predicted drug-target interactions for 30 FDA-approved drugs.

Low-Rank Embedding for Robust Image Feature Extraction.

Robustness to noises, outliers, and corruptions is an important issue in linear dimensionality reduction. Since the sample-specific corruptions and outliers exist, the class-special structure or the local geometric structure is destroyed, and thus, many existing methods, including the popular manifold learning- based linear dimensionality methods, fail to achieve good performance in recognition tasks. In this paper, we focus on the unsupervised robust linear dimensionality reduction on corrupted data by introducing the robust low-rank representation (LRR). Thus, a robust linear dimensionality reduction technique termed low-rank embedding (LRE) is proposed in this paper, which provides a robust image representation to uncover the potential relationship among the images to reduce the negative influence from the occlusion and corruption so as to enhance the algorithm's robustness in image feature extraction. LRE searches the optimal LRR and optimal subspace simultaneously. The model of LRE can be solved by alternatively iterating the argument Lagrangian multiplier method and the eigendecomposition. The theoretical analysis, including convergence analysis and computational complexity, of the algorithms is presented. Experiments on some well-known databases with different corruptions show that LRE is superior to the previous methods of feature extraction, and therefore, it indicates the robustness of the proposed method. The code of this paper can be downloaded from http://www.scholat.com/laizhihui.

Pose-Dependent Low-Rank Embedding for Head Pose Estimation

Head pose estimation via embedding model has beendemonstrated its effectiveness from the recent works.However, most of the previous methods only focuson manifold relationship among poses, while overlookthe underlying global structure among subjects and poses.To build a robust and effective head pose estimator,we propose a novel Pose-dependent Low-Rank Embedding(PLRE) method, which is designed to exploita discriminative subspace to keep within-pose samplesclose while between-pose samples far away. Specifically,low-rank embedding is employed under the multitaskframework, where each subject can be naturallyconsidered as one task. Then, two novel terms are incorporatedto align multiple tasks to pursue a better posedependentembedding. One is the cross-task alignmentterm, aiming to constrain each low-rank coefficient toshare the similar structure. The other is pose-dependentgraph regularizer, which is developed to capture manifoldstructure of same pose cross different subjects. Experimentson databases CMU-PIE, MIT-CBCL, and extendedYaleB with different levels of random noise areconducted and six embedding model based baselinesare compared. The consistent superior results demonstratethe effectiveness of our proposed method.

Bilinear low-rank coding framework and extension for robust image recovery and feature representation

We mainly study the low-rank image recovery problem by proposing a bilinear low-rank coding framework called Tensor Low-Rank Representation. For enhanced low-rank recovery and error correction, our method constructs a low-rank tensor subspace to reconstruct given images along row and column directions simultaneously by computing two low-rank matrices alternately from a nuclear norm minimization problem, so both column and row information of data can be effectively preserved. Our bilinear approach seamlessly integrates the low-rank coding and dictionary learning into a unified framework. Thus, our formulation can be treated as enhanced Inductive Robust Principal Component Analysis with noise removed by low-rank representation, and can also be considered as the enhanced low-rank representation with a clean informative dictionary via low-rank embedding. To enable our method to include outside images, the out-of-sample extension is also presented by regularizing the model to correlate image features with the low-rank recovery of the images. Comparison with other criteria shows that our model exhibits stronger robustness and enhanced performance. We also use the outputted bilinear low-rank codes for feature learning. Two unsupervised local and global low-rank subspace learning methods are proposed for extracting image features for classification. Simulations verified the validity of our techniques for image recovery, representation and classification.

An Edge-based Clustering Algorithm to Detect Social Circles in Ego Networks

Organizing users’ friends in personal social networks, i.e., ego networks, into circles is an important task for online social networks. Social networking sites allow users to manually categorize their friends into social circles. However, it is time consuming and does not update automatically as a user adds more friends. In this paper, we propose an edge-based clustering algorithm to detect social circles in ego networks automatically. Firstly, we reconstruct ego networks by predicting the missing links. Then, we define the similarity of adjacent edges and cluster edges by single-linkage hierarchical clustering algorithm. Finally, we label each circle by abstracting its common properties to explain why this circle forms. The experimental results demonstrate it is a better way to characterize social circles from the respect of edges. Our algorithm outperforms the link community algorithm and low-rank embedding algorithm in terms of accuracy, and is more efficient than the probabilistic model algorithm. Our proposed method is validated as an effective algorithm in identifying social circles.