Toward an ImageNet Library of Functions for Global Optimization Benchmarking

TL;DR

This paper introduces a novel approach to characterize BlackBox Optimization problems by transforming landscape analysis into an image recognition task, enabling automated feature extraction and benchmarking.

Contribution

It proposes converting landscape features into images and applying neural networks for classification, paving the way for an ImageNet-like function library for optimization benchmarking.

Findings

Successful classification of benchmark functions using neural networks.

Demonstrated potential for automated, conception-free landscape feature detection.

Validated approach on BBOB and IOHprofiler suites.

Abstract

Knowledge of search-landscape features of BlackBox Optimization (BBO) problems offers valuable information in light of the Algorithm Selection and/or Configuration problems. Exploratory Landscape Analysis (ELA) models have gained success in identifying predefined human-derived features and in facilitating portfolio selectors to address those challenges. Unlike ELA approaches, the current study proposes to transform the identification problem into an image recognition problem, with a potential to detect conception-free, machine-driven landscape features. To this end, we introduce the notion of Landscape Images, which enables us to generate imagery instances per a benchmark function, and then target the classification challenge over a diverse generalized dataset of functions. We address it as a supervised multi-class image recognition problem and apply basic artificial neural network models to solve it. The efficacy of our approach is numerically validated on the noise free BBOB and IOHprofiler benchmarking suites. This evident successful learning is another step toward automated feature extraction and local structure deduction of BBO problems. By using this definition of landscape images, and by capitalizing on existing capabilities of image recognition algorithms, we foresee the construction of an ImageNet-like library of functions for training generalized detectors that rely on machine-driven features.