Code covered by the BSD License  

Highlights from
Many testfunctions for global optimizers

image thumbnail
from Many testfunctions for global optimizers by Rody Oldenhuis
test functions for global optimizers

mccormick(X)
function varargout = mccormick(X)
% McCormick function 
%
%   MCCORMICK([x1, x2]) returns the value of the value of the McCormick
%   function at the specified points. [x1] and [x2] may be vectors.
%   The search domain is
%
%               -1.5 < x_1 < 4
%                -3  < x_2 < 4
%
%   The global minimum is 
%
%               f(x1, x2) = f(-0.54719, -1.54719) = -1.9133.

% Author: Rody P.S. Oldenhuis
% Delft University of Technology
% E-mail: oldenhuis@dds.nl
% Last edited 20/Jul/2009

    % if no input is given, return dimensions, bounds and minimum
    if (nargin == 0)
        varargout{1} = 2;  % # dims
        varargout{2} = [-1.5, -3]; % LB
        varargout{3} = [+4, +4]; % UB
        varargout{4} = [-5.471975602214493e-001, -1.547197559268372e+000]; % solution
        varargout{5} = -1.913222954981037e+000; % function value at solution

    % otherwise, output function value
    else
    
    % split input vector X into x1, x2        
    if size(X, 1) == 2
        x1 = X(1, :);        x2 = X(2, :); 
    else        
        x1 = X(:, 1);        x2 = X(:, 2);        
    end
    
    %   keep all values in the search domain
    x1(x1 < -1.5) = inf;       x1(x1 > 4) = inf;
    x1(x2 < -3.0) = inf;       x2(x2 > 4) = inf;
    
    % output function value
    varargout{1} = sin(x1 + x2) + (x1-x2).^2 - 1.5*x1 + 2.5*x2 + 1;
    
    end
     
end

Contact us at files@mathworks.com