Unsupervised feature selection (UFS) aims to remove the redundant information and select the most representative feature subset from the original data, so it occupies a core position for high-dimensional data preprocessing. Many proposed approaches use self-expression to explore the correlation between the data samples or use pseudolabel matrix learning to learn the mapping between the data and labels. Furthermore, the existing methods have tried to add constraints to either of these two modules to reduce the redundancy, but no prior literature embeds them into a joint model to select the most representative features by the computed top ranking scores. To address the aforementioned issue, this article presents a novel UFS method via a convex non-negative matrix factorization with an adaptive graph constraint (CNAFS). Through convex matrix factorization with adaptive graph constraint, it can dig up the correlation between the data and keep the local manifold structure of the data. To our knowledge, it is the first work that integrates pseudo label matrix learning into the self-expression module and optimizes them simultaneously for the UFS solution. Besides, two different manifold regularizations are constructed for the pseudolabel matrix and the encoding matrix to keep the local geometrical structure. Eventually, extensive experiments on the benchmark datasets are conducted to prove the effectiveness of our method. The source code is available at: https://github.com/misteru/CNAFS.

Vision-to-language tasks aim to integrate computer vision and natural language processing together, which has attracted the attention of many researchers. For typical approaches, they encode image into feature representations and decode it into natural language sentences. While they neglect high-level semantic concepts and subtle relationships between image regions and natural language elements. To make full use of these information, this paper attempt to exploit the text-guided attention and semantic-guided attention (SA) to find the more correlated spatial information and reduce the semantic gap between vision and language. Our method includes two-level attention networks. One is the text-guided attention network which is used to select the text-related regions. The other is SA network which is used to highlight the concept-related regions and the region-related concepts. At last, all these information are incorporated to generate captions or answers. Practically, image captioning and visual question answering experiments have been carried out, and the experimental results have shown the excellent performance of the proposed approach.

Unsupervised feature selection can play an important role in addressing the issue of processing massive unlabelled high-dimensional data in the domain of machine learning and data mining. This paper presents a novel unsupervised feature selection method, referred to as Multi-Group Adaptive Graph Representation (MGAGR). Different from existing methods, the relationship between features is explored via the global similarity matrix, which is reconstructed by local similarities of multiple groups. Specifically, the similarity of a feature compared to other features can be represented by the linear combination of all the local similarities. The local similarity of a representative group is given a large weight to reconstruct the global similarity. Besides, an iterative algorithm is given to solve the optimization problem, in which the global similarity matrix, its corresponding reconstruction weights and the self-representation matrix are iteratively improved. Experimental results on 8 benchmark datasets demonstrates that the proposed method outperforms the state-of-the-art unsupervised feature selection methods in terms of clustering performance. The source code is available at: https://github.com/misteru/MGAGR.

Band selection (BS) is an effective dimensionality reduction technique for hyperspectral images. Although many relevant methods have been proposed, they often only focus on the bandwise information and the correlation between the bands, and few of them pay attention to the manifold preservation in low-dimensional space, which may lead to the intrinsic structure of the data being damaged. In this article, we propose a novel method called sparse principal component analysis and adaptive multigraph learning (SPCA-AMGL) to address this issue. First, it applies SPCA to the BS task to learn the projection weight matrix, which utilizes the orthogonal constraint to remove redundant bands and uses the $ {L}_{\text{2,1}}$ norm to impose a sparse regularization on the weight matrix to ensure the selection of effective bands. Then, to select the bands with manifold preserving capability, AMGL is proposed to capture the local neighbor structure of data by combining the benefit of multiple graphs, which can not only adaptively learn the graph structure but also obtain the analytical solution of the multigraph coefficients. Finally, an alternate iterative algorithm is designed to optimize the proposed method. Abundant experiments on three hyperspectral datasets prove the reliability and superiority of the proposed method.

In the past era of explosive data growth, how to deal with large-scale, unlabeled remote sensing images (RSI) has become a concern. Due to the lack of data labels, unsupervised methods are usually used to deal with them. As one of the best unsupervised algorithms, spectral clustering (SC) is a very effective data processing and analysis technology. However, the scalability of SC affects its development in the era of big data. Deep network technology has developed rapidly in the past, and it is often used to solve the problem of out-of-sample expansion (OOSE) that traditional machine learning can’t solve. However, because RSI is usually high-dimensional data, it is easy to cause dimension explosion in deep networks. The main motivation of this work is to solve the above problems. We have designed a new algorithm called Deep Spectral Clustering with Projected Adaptive Feature Selection (DSCFS), which benefits from deep learning theory and feature projection technology. We use neural network to map RSI data, and then further process the data through regularization embedding. At the same time, adaptive feature projection is applied to extract the main features of RSI, and the loss feedback network is calculated through these two steps. A large number of experiments show that the performance of our proposed method is better than other mainstream methods.