A dimensional-level adaptive differential evolutionary algorithm for continuous optimization

In differential evolution (DE), the optimal value of the control parameters are problem-dependent. Many improved DE algorithms have been proposed with the aim of improving the exploration ability by adaptively adjusting the values of F. In those algorithms, although the value of F is adaptive at the individual level or at the population level, the value is the same for all dimensions of each individual. Individuals are close to the global optimum at some dimensions, but they may be far away from the global optimum at other dimensions. This indicated that different values of F may be needed for different dimensions. This paper proposed an adaptive scheme for the parameter F at the dimensional level. The scheme was incorporated into the jDE algorithm and tested on a set of 25 scalable benchmark functions. The results showed that the scheme improved the performance of the jDE algorithm, particularly in comparisons with several other peer algorithms on high-dimensional functions.