Robust Nonnegative Matrix Factorization via Joint Graph Laplacian and Discriminative Information for Identifying Differentially Expressed Genes

Algorithm 1

GLD-RNMF.

Input: Data matrix , indicator matrix , parameters .

Output: Matrices , and .

Initialization: Randomly initialize three nonnegative matrices ,

    and , initialize , to be identity

   matrix. Set

Repeat

   Update , and separately by

   Calculate the diagonal matrices and by (11) and (12), separately.

Until convergence.