RCMAES: A Robust CMA-ES Variant for CEC2026 Competition
For researchers in black-box optimization, this paper presents an incremental improvement to CMA-ES that enhances robustness across multiple benchmark suites.
RCMAES, a new CMA-ES variant with dimension-dependent population-size reduction and adaptive restart, achieves competitive and robust performance on CEC2017, CEC2020, and CEC2022 benchmarks compared to state-of-the-art DE algorithms and BIPOP-aCMAES.
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.