An Improved Robust Principal Component Analysis Model for Anomalies Detection of Subway Passenger Flow

Algorithm 1

Solving for the improved RPCA.

Input: Data matrix , the parameters ,

Initialize:  ,

, ,

,

, , , , .

1:while  not converged and   do

2:Update via (9)

3:Update via (11)

4:Update via (12)

5:Update via (13)

6:Update via (14)

7:Update the multipliers: via (15)

8:.

Output: Expected matrix , Sparse matrix .