Dimensionality Reduction with Sparse Locality for Principal Component Analysis

Algorithm 2

SLPCA procedure.

	Input: Training set X = {x1, x2, …, xN}, the reduced dimension k (with k ≤ D).
	Step 1: Select parameters: ρ, θ, λ1, λ2.
	Step 2: Alternative Optimization
	Initialize ρ = 0, θ = 0, U = 0, V = 0.
	While no convergence do
	(1) Fix the other variables and update U using equation (20);
	(2) Fix the other variables and update using equation (22);
	(3) Fix the other variable and update the auxiliary using equation (25);
	(4) Fix the other variables and update W using equation (27);
	(5) Update the Lagrange multiplier and other parameters using equation (28);
	End and
	Output: k-dimensional vector U, , , .