A self-supervised feature selection algorithm for spectral embedding based on block HSIC lasso

Abstract With the rapid development of information technology, a large amount unlabeled high-dimensional data has been generated. To be able to better handle these data, we propose new self-supervised feature selection algorithm for spectral embedding based on block HSIC lasso (FSSBH). It innovatively applies theoretical approach scenarios importance assessment, and performs by learning with pseudo-label matrix formed embedding. The is compared five popular algorithms classification experiments seven publicly available datasets. results show superior performance FSSBH algorithm.