A Review of Population-Based Metaheuristics for Large-Scale Black-Box Global Optimization—Part I

Scalability of optimization algorithms is a major challenge in coping with the ever-growing size problems wide range application areas from high-dimensional machine learning to complex large-scale engineering problems. The field global concerned improving scalability algorithms, particularly, population-based metaheuristics. Such metaheuristics have been successfully applied continuous, discrete, or combinatorial ranging several thousand dimensions billions decision variables. In this two-part survey, we review recent studies black-box help researchers and practitioners gain bird’s-eye view field, learn about its trends, state-of-the-art algorithms. Part I series covers two algorithmic approaches optimization: 1) problem decomposition 2) memetic II other optimization, describes areas, finally, touches upon pitfalls challenges current research identifies potential for future research.