(1) Choose 𝐾 , the number of principal axes or eigenvectors required to estimate. Consider matrix 𝐓 and set
    𝑝 ← 1 .
(2) Initialize eigenvector 𝐮 𝑝 of size 𝑑 × 1 , for example, randomly;
(3) Update 𝐮 𝑝 as 𝐮 𝑝 ← 𝐓 𝐮 𝑝 ;
(4) Do the Gram-Schmidt orthogonalization process 𝐮 𝑝 ← 𝐮 𝑝 − ∑ 𝑗 = 𝑝 − 1 𝑗 = 1 ( 𝐮 𝑇 𝑝 𝐮 𝑗 ) 𝐮 𝑗 ;
(5) Normalize 𝐮 𝑝 by dividing it by its norm: 𝐮 𝑝 ← 𝐮 𝑝 / ‖ 𝐮 𝑝 ‖ .
(6) If 𝐮 𝑝 has not converged, go back to step 3.
(7) Increment counter 𝑝 ← 𝑝 + 1 and go to step 2 until 𝑝 equals 𝐾 .
Algorithm 2: Fixed-point