Lu Q, Zhou S, Tao F, Luo J, Wang ZG.

Received: 23 April 2019 / Accepted: 18 August 2020.

Abstract

When solving a symbolic regression problem, the gene expression programming (GEP)algorithm could fall into a premature convergence which terminates the optimization pro-cess too early, and may only reach a poor local optimum. To address the premature conver-gence problem of GEP, we propose a novel algorithm named SPJ-GEP, which can maintainthe GEP population diversity and improve the accuracy of the GEP search by allowing thepopulation to jump efficiently between segmented subspaces. SPJ-GEP first divides thespace of mathematical expressions intoksubspaces that are mutually exclusive. It thencreates a subspace selection method that combines the multi-armed bandit and the-greedy strategy to choose a jump subspace. In this way, the analysis is made on the pop-ulation diversity and the range of the number of subspaces. The analysis results show thatSPJ-GEP does not significantly increase the computational complexity of time and spacethan classical GEP methods. Besides, an evaluation is conducted on a set of standard SRbenchmarks. The evaluation results show that the proposed SPJ-GEP keeps a higher popu-lation diversity and has an enhanced accuracy compared with three baseline GEP methods.

More info about thie paper

Please see Sciencedirect-Link

Code Link

Please see Code on our GitHub.