English

CMA-ES with Learning Rate Adaptation

Neural and Evolutionary Computing 2024-09-30 v2 Optimization and Control

Abstract

The covariance matrix adaptation evolution strategy (CMA-ES) is one of the most successful methods for solving continuous black-box optimization problems. A practically useful aspect of the CMA-ES is that it can be used without hyperparameter tuning. However, the hyperparameter settings still have a considerable impact on performance, especially for difficult tasks, such as solving multimodal or noisy problems. This study comprehensively explores the impact of learning rate on the CMA-ES performance and demonstrates the necessity of a small learning rate by considering ordinary differential equations. Thereafter, it discusses the setting of an ideal learning rate. Based on these discussions, we develop a novel learning rate adaptation mechanism for the CMA-ES that maintains a constant signal-to-noise ratio. Additionally, we investigate the behavior of the CMA-ES with the proposed learning rate adaptation mechanism through numerical experiments, and compare the results with those obtained for the CMA-ES with a fixed learning rate and with population size adaptation. The results show that the CMA-ES with the proposed learning rate adaptation works well for multimodal and/or noisy problems without extremely expensive learning rate tuning.

Keywords

Cite

@article{arxiv.2401.15876,
  title  = {CMA-ES with Learning Rate Adaptation},
  author = {Masahiro Nomura and Youhei Akimoto and Isao Ono},
  journal= {arXiv preprint arXiv:2401.15876},
  year   = {2024}
}

Comments

Accepted for ACM TELO. arXiv admin note: text overlap with arXiv:2304.03473

R2 v1 2026-06-28T14:29:42.845Z