Heterogeneous differential evolution for numerical optimization.

Area of Science:

• Optimization algorithms
• Computational intelligence
• Evolutionary computation

Background:

• Differential evolution (DE) is a robust population-based stochastic search algorithm.
• Standard DE and its variants typically use a single search strategy, limiting adaptability.
• Balancing exploration and exploitation is crucial for effective optimization.

Purpose of the Study:

• To propose a simple and effective heterogeneous DE (HDE) algorithm.
• To enhance the adaptability of DE by allowing individuals to adopt different search behaviors.
• To improve the balance between exploration and exploitation in optimization.

Main Methods:

• Developed a heterogeneous DE (HDE) framework where individuals randomly select search strategies from a pool.
• Implemented HDE with a pool of DE schemes to introduce diversity in population search.
• Conducted experiments on a comprehensive set of benchmark functions, including classical and shifted large-scale problems.

Main Results:

• HDE demonstrated promising performance across a majority of the tested benchmark functions.
• The heterogeneous approach led to a better balance between exploration and exploitation.
• Experimental results indicate improved effectiveness compared to standard DE on complex problems.

Conclusions:

• Heterogeneous DE (HDE) offers a simple yet effective approach to enhance optimization.
• Allowing individuals to follow diverse search behaviors improves algorithm adaptability.
• HDE shows significant potential for solving complex optimization problems.