A mapping-based constraint-handling technique for evolutionary algorithms with its applications to portfolio optimization problems

A novel Constraint-Handling Technique (CHT) for Evolutionary Algorithms (EAs) applied to constrained optimization problems is proposed. It assumed that the feasible region of problem defined by a convex-hull multiple vertices. On other hand, without loss generality, search space EA given hyper-cube. The proposed CHT called Convex-Hull Mapping (CHM) transforms real vector in into solution region. also proven CHM performs surjective mapping from Although can be any EAs, one latest or Adaptive Differential Evolution (ADE), used this paper. By using ADE, compared with conventional CHTs real-world field finance, namely portfolio problem. Portfolio process determining best proportion investment different assets according some objective. Specifically, reveal characteristic depending on number above vertices, three formulations are employed evaluate performance ADE CHM. Numerical experiments show better than most cases. Moreover, hybrid method combining outperforms original CHT.