In this paper, we propose a non-convex formulation to recover the authentic structure from the corrupted real data. Typically, the specific structure is assumed to be low rank, which holds for a wide range of data, such as images and videos. Meanwhile, the corruption is assumed to be sparse. In the literature, such a problem is known as Robust Principal Component Analysis (RPCA), which usually ...

Previous studies show that existing speech enhancement algorithms can improve speech quality but not speech intelligibility. In this study, we propose a modulation subspace (MS) based speech enhancement framework, in which the spectrogram of noisy speech is decoupled as the product of a spectral envelop subspace and a spectral details subspace. This decoupling approach provides a method to spec...

In this note we introduce a method for robust principal component regression. Robust principal components are computed from the predictor variables, and they are used afterwards for estimating a response variable by performing robust linear multiple regression. The performance of the method is evaluated at a test data set from geochemistry. Then it is used for the prediction of censored values ...

Functional principal component analysis is essential in functional data analysis, but the inference will become unconvincing when non-Gaussian characteristics occur (e.g., heavy tail and skewness). The focus of this manuscript to develop a robust methodology deal with longitudinal data, where sparsity irregularity along non-negligible measurement errors must be considered. We introduce Kendall’...

In multiple regression, different techniques are available to deal with the situation where predictors large in number, and multicollinearity exists among them. Some of these approaches rely on correlation others depend principal components. To cope influential observations (outliers, leverage, or both) data matrix for regression purposes, two proposed this paper. These Robust Correlation Based...