Table 5 The top 5 important factors selected by BART for the maximum bulk modulus K in MAX phases with additional non-informative features with the initial set of sample size N ∈ {2, 5, 10, 15, 20}.
From: Bayesian optimization with adaptive surrogate models for automated experimental design
Setting | Top 1 | Top 2 | Top 3 | Top 4 | Top 5 |
|---|---|---|---|---|---|
N = 2 | APF | ColA | ColM | Idist | \(\frac{e}{a}\) |
N = 5 | APF | ColA | Idist | ColM | \(\frac{e}{a}\) |
N = 10 | APF | ColM | ColA | Idist | \(\frac{e}{a}\) |
N = 15 | APF | ColA | Idist | ColM | \(\frac{e}{a}\) |
N = 20 | APF | ColA | ColM | Idist | \(\frac{e}{a}\) |
