An algorithm selection dataset for 1120 five-dimensional MA-BBOB test problems. It uses Exploratory Landscape Analysis (ELA) features to learn which of five candidate 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 algorithm selection based on problem landscape features.
- Benchmarking algorithm performance across a large set of optimization problems.
- Developing meta-learning approaches for black-box optimization.
- Analyzing the relationship between ELA features and algorithm performance.
Strengths
- Focuses on a well-defined algorithm selection problem with 1120 test instances.
- Uses a systematic feature set (ELA) designed for characterizing optimization landscapes.
- Has a clear, permissive license (CC-BY-4.0).
Limitations
- Row count and column details are unknown, which may limit suitability assessment.
- Column-level documentation is absent; field semantics must be inferred after download.
- Last update date is unknown; freshness unverified.
Provenance
- Source
- OpenML platform
- Collection Method
- Likely generated from benchmarking runs on MA-BBOB test suites.
- Time Range
- null
- Freshness
- null
- Geography
- null