CEC2022 Benchmark: Optimal Values and Standard Deviations for 7 Algorithms

Numerical results from 30 runs of 7 evolutionary algorithms on 12 test functions across 10 dimensions. The data includes average optimal values and standard deviations, compiled by Yang Cao. The dataset was last updated on 2026-05-13.

Use Cases

• Benchmarking algorithm performance based on average optimal values and standard deviations.
• Comparing the robustness of different algorithms (SaDE, SHADE, ILSHADE, jSO, MPEDE, LSHADE, RL-DE) across multiple test functions.
• Analyzing the statistical significance of results from 30 independent runs per algorithm and function.
• Validating new optimization methods against established CEC2022 benchmark functions.

Strengths

• Results are aggregated from 30 independent runs per algorithm, providing a basis for statistical analysis.
• Covers 7 distinct evolutionary algorithms, allowing for direct comparison.
• Uses the standardized CEC2022 benchmark suite of 12 test functions.
• Data is shared under a permissive CC-BY-4.0 license.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The dataset is very small (10.7 KB), indicating limited scope, likely containing only summary statistics.

Provenance

Source

Yang Cao via figshare

Collection Method

Likely generated from computational experiments running algorithms on benchmark functions.

Freshness

Last updated 2026-05-13 17:44:42; freshness should be verified.