Saved in:
Bibliographic Details
Main Authors: Gjorgjieva, Sara, Tuba, Eva, Seljak, Barbara Koroušić, Doerr, Carola, Eftimov, Tome
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2605.28121
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911723732074496
author Gjorgjieva, Sara
Tuba, Eva
Seljak, Barbara Koroušić
Doerr, Carola
Eftimov, Tome
author_facet Gjorgjieva, Sara
Tuba, Eva
Seljak, Barbara Koroušić
Doerr, Carola
Eftimov, Tome
contents Landscape feature representations play a central role in automated algorithm selection and meta-learning for black-box optimization, yet little is known about how different representations agree (or disagree) in the structures they impose on problem spaces. This paper presents a systematic unsupervised evaluation of four state-of-the-art representations (ELA, DeepELA, TransOptAS, and DoE2Vec) using a diverse set of affine combinations of BBOB functions (MA-BBOB). By applying extensive clustering analyses, coverage-based stability measures, and cross-representation similarity assessments, we show that each representation organizes the same problems in markedly different ways: ELA and TransOptAS form compact geometric structures, DeepELA provides a balanced intermediate view, and DoE2Vec achieves strong semantic alignment but with substantial fragmentation. Our results reveal that no single representation dominates; rather, they capture complementary aspects of the underlying landscapes. These findings highlight the importance of multi-view analyses for understanding representation behavior and offer guidance on selecting or combining representations in downstream meta-learning and algorithm selection tasks. In addition, across two different algorithm families (Differential Evolution and Particle Swarm Optimization), we show that landscape representations face an inherent trade-off in how well they align structural landscape descriptions with observed performance, indicating that no single representation can fully capture algorithm performance.
format Preprint
id arxiv_https___arxiv_org_abs_2605_28121
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle On the Structural (Dis)Agreement of Landscape Representations in Black-Box Optimization
Gjorgjieva, Sara
Tuba, Eva
Seljak, Barbara Koroušić
Doerr, Carola
Eftimov, Tome
Neural and Evolutionary Computing
Landscape feature representations play a central role in automated algorithm selection and meta-learning for black-box optimization, yet little is known about how different representations agree (or disagree) in the structures they impose on problem spaces. This paper presents a systematic unsupervised evaluation of four state-of-the-art representations (ELA, DeepELA, TransOptAS, and DoE2Vec) using a diverse set of affine combinations of BBOB functions (MA-BBOB). By applying extensive clustering analyses, coverage-based stability measures, and cross-representation similarity assessments, we show that each representation organizes the same problems in markedly different ways: ELA and TransOptAS form compact geometric structures, DeepELA provides a balanced intermediate view, and DoE2Vec achieves strong semantic alignment but with substantial fragmentation. Our results reveal that no single representation dominates; rather, they capture complementary aspects of the underlying landscapes. These findings highlight the importance of multi-view analyses for understanding representation behavior and offer guidance on selecting or combining representations in downstream meta-learning and algorithm selection tasks. In addition, across two different algorithm families (Differential Evolution and Particle Swarm Optimization), we show that landscape representations face an inherent trade-off in how well they align structural landscape descriptions with observed performance, indicating that no single representation can fully capture algorithm performance.
title On the Structural (Dis)Agreement of Landscape Representations in Black-Box Optimization
topic Neural and Evolutionary Computing
url https://arxiv.org/abs/2605.28121