ACS4Fun: An Ant Colony System for General Function Optimization

This is a C++ code that was developed to find the solution(s) that minimize any given general function of real variables. It was first developed in 2003 and a related paper was published in 2004 (in Chinese). Although it is an old code and the performance of the algorithm is not as good as the state-of-the-art algorithms, it could still be used by people with a non-optimization background to solve many function optimization problems.

The main.cpp contains a simple example to use the class CAcs4Fun to find the optimal solution of function

[y = sum{(x_i-0.5)^2}]

This function is defined in the function evaluate() as the default objective function. We can modify the definition to reflect the actual function we want to optimize (Or in a more C++ way, define evaluate() as a virtual function and then override it). However, in

CACS4Fun *acs;
acs = new CACS4Fun();
acs->alpha = 0.8;
#=> define