Evolutionary Algorithms for Large‐Scale Multi‐Objective Optimization

This chapter introduces four representative multi‐objective evolutionary algorithms (MOEAs) for solving large scale multi‐objective optimization problems (LSMOPs). The most significant feature of these MOEAs is that they suggest different ideas to handle the high‐dimensional decision spaces, such as the decision variable grouping methods in cooperative coevolution generalized differential evolution 3 (CCGDE3) and large‐scale many‐objective evolutionary algorithm. Decision variable grouping‐based MOEAs are good at solving LSMOPs with few variable interactions, decision space reduction‐based MOEAs can quickly converge to local optimums of LSMOPs with simple Pareto sets, and novel search strategy‐based MOEAs are more robust in solving complex LSMOPs. Besides, most large‐scale MOEAs are ineffective for solving small‐scale problems.