Search

Most of the existing classification systems assume that the data used is high-quality labeled. However, the labeling process in real-world may inevitably introduce corruptions into labels which can confuse the performances of classifiers. In this paper, based on Broad Learning System (BLS), we propose a novel label noise tolerant method to classify the pattern with corrupted labels. The standard BLS has shown promising efficiency and accuracy in general classification, but its learning process is prone to be affected by the noisy labels. Here, by detailed probabilistic analysis, we first give the reason for lacks of robustness in standard BLS. Then a maximum likelihood estimation-based objective function is derived for robust classification. In addition, a manifold regularization term is integrated to preserve the local geometry of data, which makes the model to be more robust and flexible to learn the output weights. Given some basic assumptions on the approximation errors, the obtained model can be transformed to a graph regularized reweighted BLS problem. The negative effects of noisy labels in data can be inhibited adaptively by assigning reasonable weights. Theoretical analysis and extensive experiments are provided to demonstrate the robustness and effectiveness of the proposed robust BLS model, especially for the case of large amounts of noisy labels. [ABSTRACT FROM AUTHOR]

The broad learning system (BLS) has garnered significant attention in the realm of visual classification due to its exceptional balance between accuracy and efficiency. However, the supervision mechanism in BLS typically relies on strict binary labels, limiting the approximation freedom and failing to represent the data distribution adequately. Furthermore, the inadequacy of the guidance mechanism for the output weights hinders the precise approximation of the target spaces by the input features. To tackle these issues, in this paper, we propose two discriminative elastic-net regularized BLS models with label enhancement, achieving the following significant objectives: Firstly, the proposed label enhancement technologies can substantially augment the margins between different classes while enhancing the diversity within label spaces. Secondly, the compactness and effectiveness of the output weights can be further improved via the guidance of elastic-net regularization of singular values. Thirdly, our proposed algorithms can be efficiently optimized with the augmented Lagrangian method, whose convergence and calculation complexity can be guaranteed well with solid theoretical analysis. Extensive experiments are intended on various popular databases to compare our proposed models with numerous other state-of-the-art recognition algorithms. The numerical results indicate that our proposed algorithms can achieve the best face recognition accuracy of 99.67%, and 97.54% for object recognition. Even in challenging recognition tasks, our methods can still yield an average improvement of 0.8 percentage points. • Label enhancement and elastic-net regularization are integrated into the BLS framework. • Two efficient optimization algorithms are devised to address our models. • Theoretical analysis support the convergence and efficiency of the optimization. • Experimental results confirm the superiority of our proposed methods. [ABSTRACT FROM AUTHOR]

Because of its simple network structure and efficient learning mode, the Broad Learning System (BLS) has achieved impressive performance in image classification tasks. Nevertheless, two deficiencies still exist which have severely limited its learning ability. First, the strict binary labeling strategy used in BLS-based models restricts the model's flexibility. Second, the final broad features are inevitably redundant, which can cause useless features to be learned and reduce the recognition accuracy. In this paper, we propose three discriminative BLS-based models to address these mentioned problems. Specifically, we first integrate the ɛ -dragging technique into the framework of standard BLS to relax the regression targets and propose the ℓ 2 -norm based discriminative BLS (L2DBLS) model. Secondly, to avoid the negative effects of redundant features in L2DBLS, we utilize the ℓ 2 , 1 regularizer to replace the Frobenius norm for feature selection. Furthermore, we propose to constrain the projection matrix of BLS by ℓ 2 and ℓ 2 , 1 regularization simultaneously. As a result, the obtained output weights can be more compact and smooth for recognition. Efficient iterative methods based on the alternating direction method of multipliers are derived to optimize the proposed models. Finally, various experiments on image databases are intended to demonstrate the outstanding recognition capability of our proposed models in comparison with other state-of-the-art classifiers. • Three regularized discriminative BLS models are proposed according to the ɛ -dragging technique and regularization theory. These models can simultaneously learn more flexible labels and compact features for better image classification. • Efficient iterative algorithms are designed to optimize the three proposed models. Moreover, a solid theoretical analysis and experimental verification are carried out to fully demonstrate its effectiveness. • Diverse experiments are conducted to validate that our three proposed models have apparent superiority for image recognition in comparison with other state-of-the-art models. [ABSTRACT FROM AUTHOR]

• ε -dragging technique was introduced into BLS to improve the discriminative ability. • Regularization is adopted to optimize network structure. • Theoretical analysis verifies the convergence of the algorithm. • Numerical simulations are conducted to verify the performance of the algorithm. This paper presents the framework of the smoothing group L 1 / 2 regularized discriminative broad learning system for pattern classification and regression. The core idea is to improve the sparseness of the standard broad learning system and improve performance on recognition and generalization. First, the ε -dragging technique is introduced into the standard broad learning system to relax regression targets and enlarge distances between categories. Then, we integrate the group L 1 / 2 regularization to optimize the network architecture to achieve sparsity. For the original group L 1 / 2 regularization, the objective function is non-convex and non-smooth, which is hard for theoretical analysis. Therefore, we propose a simple and effective smoothing technique, i.e.,smoothing group L 1 / 2 regularization, which can effectively eliminate the deficiency of the original group L 1 / 2 regularization. As a result, the final weights projection matrix has a compact form and shows discriminative power capability. In addition, the alternating direction method of multipliers was adopted to optimize the algorithm. The simulation results show that the proposed algorithm has redundancy control capability and improved performance on recognition and generalization. The simulation results proves the efficiency of the theoretical analysis. [ABSTRACT FROM AUTHOR]