Evolving dynamic fitness measures for genetic programming

作者：

Highlights：

• Further research is conducted on dynamic fitness measure genetic programming.

• Meta genetic programming is used to evolve the dynamic fitness measures.

• Meta genetic programming approach outperforms the previous approach.

• Dynamic fitness measure genetic programming outperforms standard genetic programming.

• Reusability of composite fitness measures produced by the new approach.

摘要

•Further research is conducted on dynamic fitness measure genetic programming.•Meta genetic programming is used to evolve the dynamic fitness measures.•Meta genetic programming approach outperforms the previous approach.•Dynamic fitness measure genetic programming outperforms standard genetic programming.•Reusability of composite fitness measures produced by the new approach.

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

论文评审过程：Received 16 September 2017, Revised 15 February 2018, Accepted 28 March 2018, Available online 25 April 2018, Version of Record 29 May 2018.

论文官网地址：https://doi.org/10.1016/j.eswa.2018.03.060