Skip to main content

Table 1 Comparison of the number of high fidelity evaluations for the GPR-Hybrid approach \(|\text{HF}_{\text{GPR-H}}|\) for different sizes of the initial training data set \(|\mathcal{T}_{\text{I}}|\), using batch size \(N_{\text{B}}=50\)

From: A blackbox yield estimation workflow with Gaussian process regression applied to the design of electromagnetic devices

 

\(|\mathcal{T}_{\text{I}}|=5\)

\(|\mathcal{T}_{\text{I}}|=10\)

\(|\mathcal{T}_{\text{I}}|=30\)

\(|\text{HF}_{\text{GPR-H}}^{\text{offline}}|\)

55

110

330

\(|\text{HF}_{\text{GPR-H}}^{\text{online}}|\)

306

226

179

\(|\text{HF}_{\text{GPR-H}}^{\text{total}}|\)

361

336

509