Search

Correntropy is a local similarity measure defined in kernel space, hence can combat large outliers in robust signal processing and machine learning. So far, many robust learning algorithms have been developed under the maximum correntropy criterion (MCC), among which, a Gaussian kernel is generally used in correntropy. To further improve the learning performance, in this paper we propose the concept of mixture correntropy, which uses the mixture of two Gaussian functions as the kernel function. Some important properties of the mixture correntropy are presented. Applications of the maximum mixture correntropy criterion (MMCC) to extreme learning machine (ELM) and kernel adaptive filtering (KAF) for function approximation and data regression are also studied. Experimental results show that the learning algorithms under MMCC can perform very well and achieve better performance than the conventional MCC based algorithms as well as several other state-of-the-art algorithms. [ABSTRACT FROM AUTHOR]

Concept factorization (CF), unlike nonnegative matrix factorization (NMF), can handle data with negative values by approximating the original data with two low-dimensional nonnegative matrices and itself. Nevertheless, existing CF-based methods continue to suffer from the two issues specified as follows: (1) Their effectiveness is reduced by the high degree of factorization freedom and the two-stage mismatch between factorization and category acquisition, and (2) their robustness drops significantly when dealing with complex noise. In response to the aforementioned issues, we propose a robust spectral-embedded bilateral orthogonal concept factorization (RSOCF) model for clustering. It constrains the factor matrices as orthogonal matrices to decrease the freedom and obtain samples' categories directly after factorization, which can significantly improve clustering effectiveness. Moreover, correntropy is introduced into RSOCF to improve its robustness to complex noise. To optimize the non-convex RSOCF model, a half-quadratic-based algorithm is devised. Numerous experiments demonstrate that RSOCF surpasses other state-of-the-art methods in terms of clustering effectiveness and robustness. • A novel robust concept factorization-based clustering model is proposed. • A fast algorithm is proposed to optimize the non-convex model. • Numerous experiments validate its clustering effectiveness and robustness. [ABSTRACT FROM AUTHOR]

Randomized autoencoder (RAE) has attracted much attention due to its strong capability of representation with fast learning speed. However, the mainstream RAEs are still designed for scalar/vector data, which inevitably destroys the structure information of tensor data. To alleviate this deficiency, a novel convolutions based matrix randomized autoencoder (MRAE) is developed for two-dimensional (2D) data in this paper, including a one-side MRAE (OMRAE) exploiting the row or column information and a double-side MRAE (DMRAE) that simultaneously extracts the row and column information by 2 parallel OMRAEs. To reduce meaningless encoded features, the within-class scatter matrix (WSI) and within-class interaction distance (WID) constraints are added into OMRAE resulting WSI-OMRAE and WID-OMRAE, respectively. To demonstrate the superiority, stacked MRAEs are embedded into hierarchical regularized least squares for one-class classification and comparisons with several state-of-the-art methods are provided. The source code would be available at https://github.com/ML-HDU/MRAE. • MRAE can learn from 2D data directly without breaking structure information. • WSI-OMRAE minimizes both reconstruction error and within-class scatter. • WID-OMRAE can learn intrinsic manifold structure with compact representation. • Adopting two-sides OMRAEs can perform row/column feature learning in parallel. [ABSTRACT FROM AUTHOR]

• A novel correntropy based semi-supervised NMF method is proposed for image clustering. • The proposed method is analysed in terms of convergence, robustness, and computational complexity. • The relationships between the proposed method and several typical NMF based methods are discussed. • Experimental results show the effectiveness and robustness of the proposed method in image clustering tasks. Nonnegative matrix factorization (NMF) is a powerful dimension reduction method, and has received increasing attention in various practical applications. However, most traditional NMF based algorithms are sensitive to noisy data, or fail to fully utilize the limited supervised information. In this paper, a novel robust semi-supervised NMF method, namely correntropy based semi-supervised NMF (CSNMF), is proposed to solve these issues. Specifically, CSNMF adopts a correntropy based loss function instead of the squared Euclidean distance (SED) in constrained NMF to suppress the influence of non-Gaussian noise or outliers contaminated in real world data, and simultaneously uses two types of supervised information, i.e., the pointwise and pairwise constraints, to obtain the discriminative data representation. The proposed method is analyzed in terms of convergence, robustness and computational complexity. The relationships between CSNMF and several previous NMF based methods are also discussed. Extensive experimental results show the effectiveness and robustness of CSNMF in image clustering tasks, compared with several state-of-the-art methods. [ABSTRACT FROM AUTHOR]

• A novel spatial attention encoder is proposed, which builds global feature insight with long-range dependence to enhance perception. • We propose a context block decoder to aggregate contextual information as channel-specific bias of the features. • A feature fusion module with parallel branches considering expectation and residual property is proposed to realize adaptive information fusion. • We build an ESD dataset with 89.70% small defects collected from real industry an evaluation benchmark. Despite the promising development of Automatic Visual Inspection (AVI) in the manufacturing industry, detecting small-sized defects with fewer pixels coverage remains a challenging problem due to its insufficient attention and lack of semantic information. Most exsiting convolutional inspection methods overlook the long-range dependence of context and lack adaptive fusion strategies to exploit heterogeneous features. To address these issues in AVI, this paper proposes a novel contextual information and spatial attention based network (CANet), which consists of two steps, namely CAblock and LaplacianFPN, for effective perception and exploitation of small defect features. Specifically, CAblock extracts semantic information with rich context by encoding spatial long-range dependence and decoding contextual information as channel-specific bias through a Spatial Attention Encoder (SAE) and a Context Block Decoder (CBD), respectively. LaplacianFPN further performs adaptive feature fusion considering both feature consistency and heterogeneity via two parallel branches. As a benchmark, a self-built Engine Surface Defects (ESD) dataset collected in real industry containing 89.70% small defects is constructed. Experimental results show that CANet achieves mAP-50 improvements of 1.5% and 4.3% compared to state-of-the-art methods on NEU-DET and ESD, which demonstrates the effectiveness of the proposed method. The code is now available at https://github.com/xiuqhou/CANet. [ABSTRACT FROM AUTHOR]

• A novel hypergraph based semi-supervised SNMF method is proposed for image clustering. • A new hypergraph based pairwise constraints propagation algorithm is presented. • The proposed method is analysed in terms of convergence, supervisory information, and computational complexity. • Extensive experiments confirm the effectiveness of the proposed method in clustering applications. Semi-supervised symmetric nonnegative matrix factorization (SNMF) has been shown to be a significant method for both linear and nonlinear data clustering applications. Nevertheless, existing SNMF-based methods only adopt a simple graph to construct the similarity matrix, and cannot fully use the limited supervised information for the construction of the similarity matrix. To overcome the drawbacks of previous SNMF-based methods, a new semi-supervised SNMF-based method called hypergraph based semi-supervised SNMF (HSSNMF), is proposed in this paper for image clustering. Specifically, HSSNMF adopts a predefined hypergraph to build a similarity matrix for capturing the high-order relationships of samples. By exploiting a new hypergraph based pairwise constraints propagation (HPCP) algorithm, HSSNMF propagates the pairwise constraints of the limited data points to the entire data points, which can make full use of the limited supervised information and construct a more informative similarity matrix. Using the multiplicative updating algorithm, a discriminative assignment matrix can then be obtained by solving the optimization problem of HSSNMF. Moreover, analyses of the convergence, supervisory information, and computational complexity of HSSNMF are presented. Finally, extensive clustering experiments have been conducted on six real-world image datasets, and the experimental results have demonstrated the superiority of HSSNMF while compared with several state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Modern technologies have been producing data with complex intrinsic structures, which can be naturally represented as two-dimensional matrices, such as gray digital images, and electroencephalography (EEG) signals. When processing these data for classification, traditional classifiers, such as support vector machine (SVM) and logistic regression, have to reshape each input matrix into a feature vector, resulting in the loss of structural information. In contrast, modern classification methods such as support matrix machine capture these structures by regularizing the regression matrix to be low-rank. These methods assume that all entities within each input matrix can serve as the explanatory features for its label. However, in real-world applications, many features are redundant and useless for certain classification tasks, thus it is important to perform feature selection to filter out redundant features for more interpretable modeling. In this paper, we tackle this issue, and propose a novel classification technique called Sparse Support Matrix Machine (SSMM), which is favored for taking both the intrinsic structure of each input matrix and feature selection into consideration simultaneously. The proposed SSMM is defined as a hinge loss for model fitting, with a new regularization on the regression matrix. Specifically, the new regularization term is a linear combination of nuclear norm and ℓ 1 norm, to consider the low-rank property and sparse property respectively. The resulting optimization problem is convex, and motivates us to propose a novel and efficient generalized forward-backward algorithm for solving it. To evaluate the effectiveness of our method, we conduct comparative studies on the applications of both image and EEG data classification problems. Our approach achieves state-of-the-art performance consistently. It shows the promise of our SSMM method on real-world applications. [ABSTRACT FROM AUTHOR]