An investigation of dynamic fitness measures for genetic programming

作者：

Highlights：

• Dynamic Fitness Measure Genetic Programming is proposed.

• The approach uses a different fitness measure at each generation.

• A genetic algorithm is used to induce the sequence of fitness measures.

• The approach outperforms standard genetic programming on benchmark tasks and on complex, real-world problems.

摘要

•Dynamic Fitness Measure Genetic Programming is proposed.•The approach uses a different fitness measure at each generation.•A genetic algorithm is used to induce the sequence of fitness measures.•The approach outperforms standard genetic programming on benchmark tasks and on complex, real-world problems.

论文关键词：Genetic programming,Genetic algorithm,Fitness

论文评审过程：Received 18 March 2017, Revised 9 August 2017, Accepted 11 August 2017, Available online 12 August 2017, Version of Record 22 September 2017.

论文官网地址：https://doi.org/10.1016/j.eswa.2017.08.022