RCMAES: A Robust CMA-ES Variant for CEC2026 Competition

About

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.

Khoirul Faiq Muzakka, Sören Möller, Martin Finsterbusch · 2026

Related benchmarks

Task	Dataset	Result
Numerical Optimization	CEC D=10 2022	Friedman Rank4.029	30
Numerical Optimization	CEC D=20 2022	Friedman Rank3.17	15
Single Objective Bound Constrained Numerical Optimization	CEC 5D 2020	Accuracy1.9	7
Single Objective Bound Constrained Numerical Optimization	CEC 10D 2020	Accuracy2.3	7
Single Objective Bound Constrained Numerical Optimization	CEC 15D 2020	Accuracy (E)0.024	7
Single Objective Bound Constrained Numerical Optimization	CEC 20D 2020	Accuracy2.4	7
Single-objective bound-constrained optimization	CEC D=10 2017	Accuracy (E)3.5	7
Single-objective bound-constrained optimization	CEC D=30 2017	Accuracy (E)0.071	7
Single-objective bound-constrained optimization	CEC D=50 2017	Accuracy12.9	7
Single-objective bound-constrained optimization	CEC D=100 2017	Accuracy (E)19	7