RCMAES: A Robust CMA-ES Variant for CEC2026 Competition
Khoirul Faiq Muzakka, Sören Möller, Martin Finsterbusch
TLDR
RCMAES is a new CMA-ES variant that uses adaptive population size reduction and restart for robust performance on CEC benchmarks.
Key contributions
- Introduces RCMAES, a novel CMA-ES variant for benchmark optimization.
- Employs a dimension-dependent nonlinear population-size reduction strategy.
- Incorporates an adaptive restart mechanism within the CMA-ES framework.
- Achieves competitive and robust performance on CEC2017, CEC2020, and CEC2022 benchmarks.
Why it matters
RCMAES offers a robust and competitive approach to global optimization challenges, particularly for complex benchmark problems. Its novel strategies for population management and restarts could significantly advance the efficiency of evolutionary algorithms. This makes it a strong contender for future optimization competitions.
Original 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.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.