Table 1 Algorithm: Bayesian optimization
[1] Input: Design space X, initial dataset D0, GPR surrogate model M, acquisition function α | |
[2] Output: Best material design x* | |
[3] Initialize GPR surrogate model M with D0, D = D0 | |
[4] while not terminated do | // Until max iterations or target performance met |
[5] Train/Update GPR surrogate model M using D | |
[6] xnext = argmax x∈X α(x; M) | // Inner optimization |
[7] ynext = f(xnext) | // Evaluate design |
[8] D = D ∪ {(xnext, ynext)} | // Update dataset with experiment results |
[9] end while | |
[10] return x* = argmax x∈D f(x) | |
