mabbob_ela_as_2d_classify: Algorithm Selection for 2D BBOB Problems Using ELA Features

A dataset for algorithm selection on 1,120 two-dimensional MA-BBOB test problems. It uses Exploratory Landscape Analysis (ELA) features to predict which of five algorithms achieves the highest Area Under the Convergence Curve (AUCC). The dataset was contributed by author Diederich Vermetten and is licensed under CC-BY-4.0.

Use Cases

• Training models for automated algorithm selection based on problem landscape features.
• Benchmarking machine learning classifiers on a structured optimization meta-problem.
• Analyzing the relationship between Exploratory Landscape Analysis (ELA) features and algorithm performance.
• Developing surrogate models to recommend algorithms for new black-box optimization problems.

Strengths

• Focuses on a specific, well-defined meta-problem with 1,120 test instances.
• Uses established Exploratory Landscape Analysis (ELA) features for algorithm selection.
• Clear objective to predict the best of five algorithms based on AUCC performance.
• Published under the permissive CC-BY-4.0 license for reuse.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment for specific modeling tasks.
• Last update date is unknown; freshness unverified.

Provenance

Source

OpenML platform, contributed by Diederich Vermetten.

Collection Method

Likely generated from benchmarking runs on the MA-BBOB test suite.

Time Range

Temporal coverage is not specified.

Freshness

Last updated date is unknown.

Geography

Spatial coverage is not applicable.