Self-Adaptive Surrogate-Assisted Covariance Matrix Adaptation Evolution Strategy

TL;DR

This paper introduces saACM-ES, an adaptive surrogate-assisted CMA-ES variant that dynamically adjusts surrogate model parameters, leading to faster optimization and improved results on benchmark problems.

Contribution

It proposes a novel online adaptation mechanism for surrogate models within CMA-ES, enhancing efficiency and scalability over existing methods.

Findings

Significant speed-up over ACM-ES and CMA-ES baselines.

Effective surrogate model quality improvement through online heuristics.

Achieved new best results on some BBOB-2012 benchmark problems.

Abstract

This paper presents a novel mechanism to adapt surrogate-assisted population-based algorithms. This mechanism is applied to ACM-ES, a recently proposed surrogate-assisted variant of CMA-ES. The resulting algorithm, saACM-ES, adjusts online the lifelength of the current surrogate model (the number of CMA-ES generations before learning a new surrogate) and the surrogate hyper-parameters. Both heuristics significantly improve the quality of the surrogate model, yielding a significant speed-up of saACM-ES compared to the ACM-ES and CMA-ES baselines. The empirical validation of saACM-ES on the BBOB-2012 noiseless testbed demonstrates the efficiency and the scalability w.r.t the problem dimension and the population size of the proposed approach, that reaches new best results on some of the benchmark problems.