Brain State Decoding Based on fMRI Using Semisupervised Sparse Representation Classifications

Algorithm 1

semiSRC-AVE.

Input: training matrix , training label , test set

( is the number of test samples).

Output: identities of

Method:

for   to

Normalize the columns of to have unit -norm.

Solve the -minimization problem:

Compute and output the identity of :

Compute the distance of :

Compute the threshold of :

  if  

           ,

  end

end