Neural Exploratory Landscape Analysis for Meta-Black-Box-Optimization

TL;DR

This paper introduces NeurELA, a neural framework that automatically profiles landscape features for Meta-Black-Box Optimization, reducing reliance on human-crafted features and enhancing algorithm adaptability and performance.

Contribution

NeurELA is a novel, end-to-end neural approach that dynamically profiles landscape features, trained with multi-task neuroevolution, improving MetaBBO performance across diverse tasks.

Findings

NeurELA outperforms traditional methods in various MetaBBO tasks.

It can be efficiently fine-tuned for specific optimization problems.

NeurELA generalizes well to unseen MetaBBO algorithms.

Abstract

Recent research in Meta-Black-Box Optimization (MetaBBO) have shown that meta-trained neural networks can effectively guide the design of black-box optimizers, significantly reducing the need for expert tuning and delivering robust performance across complex problem distributions. Despite their success, a paradox remains: MetaBBO still rely on human-crafted Exploratory Landscape Analysis features to inform the meta-level agent about the low-level optimization progress. To address the gap, this paper proposes Neural Exploratory Landscape Analysis (NeurELA), a novel framework that dynamically profiles landscape features through a two-stage, attention-based neural network, executed in an entirely end-to-end fashion. NeurELA is pre-trained over a variety of MetaBBO algorithms using a multi-task neuroevolution strategy. Extensive experiments show that NeurELA achieves consistently superior performance when integrated into different and even unseen MetaBBO tasks and can be efficiently fine-tuned for further performance boost. This advancement marks a pivotal step in making MetaBBO algorithms more autonomous and broadly applicable. The source code of NeurELA can be accessed at https://github.com/GMC-DRL/Neur-ELA.