Aligned deep neural network for integrative analysis with high-dimensional input.

Journal: Journal of biomedical informatics
Published Date:

Abstract

OBJECTIVE: Deep neural network (DNN) techniques have demonstrated significant advantages over regression and some other techniques. In recent studies, DNN-based analysis has been conducted on data with high-dimensional input such as omics measurements. In such analysis, regularization, in particular penalization, has been applied to regularize estimation and distinguish relevant input variables from irrelevant ones. A unique challenge arises from the "lack of information" attributable to high dimensionality of input and limited size of training data. For many data/studies, there exist other data/studies that may be relevant and can potentially provide additional information to boost performance.

Authors

  • Shunqin Zhang
    School of Mathematical Sciences, University of Chinese Academy of Sciences, Beijing, China; Key Laboratory of Big Data Mining and Knowledge Management, Chinese Academy of Sciences, Beijing, China; Department of Biostatistics, Yale University, New Haven, CT, USA.
  • Sanguo Zhang
    c School of Mathematical Science, University of Chinese Academy of Sciences , Beijing , China.
  • Huangdi Yi
    Department of Biostatistics, Yale University, New Haven, CT, USA.
  • Shuangge Ma
    Department of Biostatistics, Yale School of Public Health, New Haven, CT 06511, USA.