Figure 2 The graph plots output value f...

Figure 2

A graph of output value f versus input variable x, showing actual points y subscript i, primary function f, and surrogate models f hat subscript 1 and f hat subscript 2 with residual areas R subscript 1 and R subscript 2.

The graph plots output value f on the vertical axis and input variable x on the horizontal axis. Discrete points represent actual training values y subscript i. A dashed curve represents the primary function f. Two smooth curves represent surrogate models f hat subscript 1 and f hat subscript 2. The curve for f hat subscript 1 lies closer to the primary function in most regions, while f hat subscript 2 deviates more. Shaded regions indicate residual areas between the primary function and each surrogate model. Residual area R subscript 1 corresponds to f hat subscript 1 and appears smaller. Residual area R subscript 2 corresponds to f hat subscript 2 and appears larger across multiple intervals.

Multiple surrogate models $\hat{f_{i}}$ can approximate the actual training values $y_{i}$ ⁠, derived from the primary function f, with similar accuracy

Sharing Unavailable