Assessing Dry Weight of Hemodialysis Patients via Sparse Laplacian Regularized RVFL Neural Network with L_2,1-Norm

Algorithm 1. Algorithm of SLapRVFL

Require: Training set , test set , the numbers of hidden layer nodes (), the maximum number of iterations tmax, coefficients of and ;

Ensure: The predictive values of

(1) Randomly initializing all weights and deviations between the hidden layer and the input layer. Calculating the hidden layer output matrix (training set)and Laplacian matrix by Equations (2), (12), and (13);

(2) Set , estimate the initial using Equation (7);

Repeat

(3) Update the diagonal matrix with

(4) Update via Equation (11d);

Until;

(5) Calculate the hidden layer output matrix (test set);

(6) Estimate by .

Algorithm 1. Algorithm of SLapRVFL