Comparison of Low-budget Black-box Optimization Algorithms on BBOB

Data replicates results from a paper submitted to IEEE Transactions on Evolutionary Computation. It compares Black-Box Optimization tools with classical heuristics on the BBOB benchmark suite from the COCO environment. The data was authored by Elena Raponi of Leiden University.

Use Cases

• Benchmarking algorithm performance based on 24 noiseless functions with different landscape characteristics.
• Comparing classical heuristics against modern black-box optimization tools for machine learning.
• Studying algorithm behavior on functions with varying properties like uni-modality, multi-modality, and separability.

Strengths

• Data is tied to a peer-reviewed paper submission to a major journal (IEEE Transactions on Evolutionary Computation).
• Benchmarking is performed on the well-known BBOB suite, which includes 24 noiseless functions with defined characteristics.

Limitations

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

Provenance

Source

Elena Raponi, Leiden University

Collection Method

Data generated to replicate results from a scientific paper.