Manifold Adaptive Kernelized Low-Rank Representation for Semisupervised Image Classification

Algorithm 1

Optimization to (18).

Input: data points , regularization parameters ,

, , , , , , and ;

Output: the low-rank representation coefficient matrix .

(1) while not converged do

(2) Fix the other variables and update by

(3) Fix the others and update by

(4) Fix the others and update by

(5) Update the multipliers

(6) Update the parameter by

(7) Check the convergence conditions

,

(8) end while