Local fitness landscape from paired comparison-based memetic search in interactive differential evolution and differential evolution

Abstract

In this paper, we propose a triple comparison-based interactive differential evolution (IDE) algorithm and a differential evolution (DE) algorithm. The comparison of target vector and trial vector supports a local fitness landscape for IDE and DE algorithms to conduct a memetic search. In addition to the target vector and trial vector used in canonical IDE and DE algorithm frameworks, we conduct a memetic search around whichever vector has better fitness. We use a random number from a normal distribution generator or a uniform distribution generator to perturb the vector, thereby generating a third vector. By comparing the target vector, the trial vector, and the third vector, we implement a triple comparison mechanism in IDE and DE algorithms. Our proposed triple comparison-based IDE and DE algorithms show significantly better optimisation performance arising from the evaluation results. We also investigate potential issues arising from our proposal and discuss some open topics and future opportunities.