RCMAES: A Robust CMA-ES Variant for CEC2026 Competition

TL;DR

This paper introduces RCMAES, a new CMA-ES variant with adaptive restart and population-size reduction, demonstrating robust performance on multiple CEC benchmark suites.

Contribution

RCMAES combines nonlinear population-size reduction and adaptive restart within CMA-ES, advancing optimization robustness and efficiency on benchmark problems.

Findings

RCMAES outperforms several state-of-the-art algorithms on CEC benchmarks.

RCMAES shows consistent robustness across diverse benchmark suites.

Experimental results validate the effectiveness of the proposed strategies.

Abstract

This paper proposes RCMAES, a novel variant of the Covariance Matrix Adaptation Evolution Strategy (CMA-ES) for CEC benchmark optimization. RCMAES integrates a dimension-dependent nonlinear population-size reduction strategy with an adaptive restart mechanism within a pure CMA-ES framework. RCMAES is evaluated on three benchmark suites (CEC2017, CEC2020, and CEC2022) and compared with state-of-the-art DE algorithms as well as its closely related counterpart, BIPOP-aCMAES. Experimental results show that RCMAES achieves competitive and robust performance across all benchmarks.