function results = simulateIncrementalLearning ( labeledData, unlabeledData, testData, b_efficientUpdates, idxToAdd )
% Copyright (c) by Alexander Freytag, 2013-11-13.

    if ( (nargin < 4) || isempty (b_efficientUpdates) )
        b_efficientUpdates = true;
    end
    
    if ( (nargin < 5) || isempty (idxToAdd) )
        idxToAdd = 1:length(unlabeledData.y);
    end    
    
    labeled_X = labeledData.X;
    labeled_y = labeledData.y;
    
    unlabeled_X = unlabeledData.X;
    unlabeled_y = unlabeledData.y;
    
    test_X = testData.X;
    test_y = testData.y;     

    % output
    ARRscore = zeros ( 1+length(unlabeled_y ) ,1 );
    times = zeros ( 1+length(unlabeled_y ) ,1 );
    
    %% (2.1) initially train the model
    
    % settings
    gpnoise = 0.1;
    cov = {'covmin'};
    loghyper = [];
 
    K = feval(cov{:},loghyper, labeled_X');
  
    timeStamp_trainStart = tic;
    model = learn_multiclass_gp(K, labeled_y', gpnoise);
    times(1) = toc(timeStamp_trainStart);
  
    
    %% (2.2) test initial model
    %copute similarities to test samples  
    Ks = feval(cov{:},loghyper, labeled_X', test_X');    
 
    [~, pred_labels] = test_multiclass_gp(model, Ks);
    
    % multi-class accuracy: 1/(# classes) \sum accuracy_per_class
    numCorrect = norm(double(pred_labels == test_y'), 1);
    testSize = double(size(test_y,1)); 
    ARRscore(1) = double(numCorrect)/double(testSize);
    
    
    %% (3) perform incremental updates
    

    
    for i_idxAdd = 1:length(unlabeled_y)
        
        %%%%%%%%%%%%%%
        %% update
        %%%%%%%%%%%%%%
        
        % get new sample to add
        xNew = unlabeled_X( :, idxToAdd(i_idxAdd) );
        % simulate asking of user
        yNew = unlabeled_y ( idxToAdd(i_idxAdd) );
        
        if ( b_efficientUpdates )
        
            % compute kernel values
            kss = feval(cov{:},loghyper, xNew');        
            ksNewSample = feval(cov{:},loghyper, labeled_X', xNew');
            

            timeStamp_updateStart = tic;
            model = update_multiclass_gp(model, kss, ksNewSample, yNew);
            times(i_idxAdd+1) = toc(timeStamp_updateStart);

            labeled_X = [ labeled_X, xNew];
        else
            kss = feval(cov{:},loghyper, xNew');        
            ksNewSample = feval(cov{:},loghyper, labeled_X', xNew'); 
            K = [K, ksNewSample;ksNewSample',kss];
            
            labeled_X = [ labeled_X, xNew];
            labeled_y = [ labeled_y; yNew];         

            timeStamp_updateStart = tic;
            model = learn_multiclass_gp(K, labeled_y', gpnoise);
            times(i_idxAdd+1) = toc(timeStamp_updateStart);
         
        end
        
        
        %%%%%%%%%%%%%
        %% test
        %%%%%%%%%%%%%
        
        Ks = [Ks; feval(cov{:},loghyper, xNew', test_X')]; 

        [~, pred_labels] = test_multiclass_gp(model, Ks);

        % multi-class accuracy: 1/(# classes) \sum accuracy_per_class
        numCorrect = norm(double(pred_labels == test_y'), 1);
        testSize = double(size(test_y,1)); 
        ARRscore(i_idxAdd+1) = double(numCorrect)/double(testSize);        
    end
    
    results.ARR = ARRscore;
    results.times = times;
end