A review of surrogate-assisted evolutionary algorithms for expensive optimization problems

Many problems in real life can be seen as Expensive Optimization Problems (EOPs). Compared with traditional optimization problems, the evaluation cost of candidate solutions for EOPs is expensive and even unaffordable. Surrogate-assisted evolutionary algorithms (SAEAs) has become a hot technology to solve EOPs in recent year, because they can effectively reduce computational cost and improve solving efficiency. However, few literatures provide a systematic overview for SAEAs. This paper systematically summarizes the existing research results of SAEAs from the aspects of algorithms and applications. Firstly, the necessity of studying SAEAs and several commonly used surrogate models are introduced. Subsequently, according to the type of objective functions and constraints, the existing SAEAs are classified and discussed. Then, the application of SAEAs in many fields are reviewed. Finally, we indicate several promising lines of research that are worthy of devotion in future.

Automated summary

This paper provides a systematic overview of surrogate-assisted evolutionary algorithms (SAEAs), including the necessity of studying SAEAs, commonly used surrogate models, classification and discussion of existing SAEAs, review of their applications in various fields, and suggestions for future research directions.