Exploratory Landscape Analysis Feature Values for the 24 Noiseless BBOB Functions

The dataset contains feature values for the so-called "cheap" features available in the R package <em>flacco</em> [https://github.com/kerschke/flacco] on the 24 noiseless BBOB functions of the COCO benchmark environment [https://github.com/numbbo/coco]. Precisely, the data set covers <strong>raw data</strong> and <strong>median</strong> feature values for the following <strong>Feature Sets: </strong><em>dispersion, \(y\)-distribution, meta model, information content, nearest better clustering</em> and <em>principal component analysis.</em> <strong>Functions: </strong>24 noiseless BBOB functions of the COCO benchmark environment [https://github.com/numbbo/coco]. <strong>Instances: </strong>First five instances of each function. <strong>Dimension \(d\): </strong>6 different dimensions, \(d \in \{5,10,15,20,25,30\}\). <strong>Sample size ("budget") \(b\) for Feature Extraction: </strong>7 different options per dimension, \(b \in \{ 30d, 50d, 100d, 250d, 650d, 800d, 1000d\}\). <strong>Sampling Distribution:</strong> Features are computed using Sobol' low-discrepancy sequence as sampling strategy. <strong>Repetitions: </strong>100 independent repetitions are available. Caused by the heavy computation times, <strong> some data records are missing:</strong> The <em>dispersion</em> feature values are not available for \((d=25,b=1000d), (d=30,b=800d) \) and \((d=30,b=1000d)\); Instances 2 to 5 are not available for: \(1000 d\) budgets with \(d \geq 10\) \(800d\) budgets with \(d \geq 15\) \(650d\) budgets with \(d \geq 20\) \(250d\) budgets with \(d = 30\)