Evolutionary Multitask Optimization in Real-World Applications: A Survey

Due to its good ability solve problems, evolutionary multitask optimization (EMTO) algorithm has been widely studied recently. Evolutionary the advantage of fast searching for optimal solution, but it is easy fall into local optimum and difficult generalize. To these an effective method combine with algorithm. Through implicit parallelism tasks themselves knowledge transfer between tasks, more promising individuals can be generated in evolution process, which jump out optimum. How better two also more. This paper will explore existing multitasking theory improvement scheme detail. Then summarizes application different scenarios. Finally, according research, future research trends potential exploration directions are revealed.